Verint ContactStore for Communication Manager Release 7.8 Planning, Installation and Administration Guide

Transcription

1 Verint ContactStore for Communication Manager Release 7.8 Planning, Installation and Administration Guide Issue 5 May, 2008

2 Verint Systems, Inc. All Rights Reserved. Notice While reasonable efforts were made to ensure that the information in this document was complete and accurate at the time of printing, Verint Systems, Inc. and Avaya Inc. can assume no liability for any errors. Changes and corrections to the information in this document may be incorporated in future releases. Preventing toll fraud "Toll fraud" is the unauthorized use of your telecommunications system by an unauthorized party (for example, anyone who is not a corporate employee, agent, subcontractor, or person working on your company's behalf). Be aware that there may be a risk of toll fraud associated with your system and that, if toll fraud occurs, it can result in substantial additional charges for your telecommunications services. Avaya fraud intervention If you suspect that you are being victimized by toll fraud and you need technical assistance or support, call Technical Service Center Toll Fraud Intervention Hotline at for the United States and Canada. Providing telecommunications security Telecommunications security (of voice, data, and video communications) is the prevention of any type of intrusion to (that is, either unauthorized or malicious access to or use of) your company's telecommunications equipment by some party. Your company's "telecommunications equipment" includes both this Avaya product and any other voice/data/ video equipment that could be accessed via this Avaya product (that is, "networked equipment"). An "outside party" is anyone who is not a corporate employee, agent, subcontractor, or person working on your company's behalf. Whereas, a "malicious party" is anyone (including someone who may be otherwise authorized) who accesses your telecommunications equipment with either malicious or mischievous intent. Such intrusions may be either to/through synchronous (time-multiplexed and/or circuit-based) or asynchronous (character-, message-, or packet-based) equipment or interfaces for reasons of: Use (of capabilities special to the accessed equipment) Theft (such as, of intellectual property, financial assets, or toll-facility access) Eavesdropping (privacy invasions to humans) Mischief (troubling, but apparently innocuous, tampering) Harm (such as harmful tampering, data loss or alteration, regardless of motive or intent) Be aware that there may be a risk of unauthorized intrusions associated with your system and/or its networked equipment. Also realize that, if such an intrusion should occur, it could result in a variety of losses to your company (including, but not limited to, human and data privacy, intellectual property, material assets, financial resources, labor costs, and legal costs). Responsibility for your company's telecommunications security The final responsibility for securing both this system and its networked equipment rests with you, an Avaya customer's system administrator, your telecommunications peers, and your managers. Base the fulfillment of your responsibility on acquired knowledge and resources from a variety of sources, including, but not limited to: Installation documents System administration documents Security documents Hardware-/software-based security tools Shared information between you and your peers Telecommunications security experts To prevent intrusions to your telecommunications equipment, you and your peers should carefully program and configure your Avaya-provided telecommunications systems and their interfaces. Your Avaya-provided software applications, as well as their underlying hardware/software platforms and interfaces. Any other equipment networked to your Avaya products. Warranty Avaya Inc. provides a limited warranty on this product. Refer to your sales agreement to establish the terms of the limited warranty. In addition, Avaya's standard warranty language, as well as information regarding support for this product, while under warranty, is available through the following Web site: Link disclaimer Avaya Inc. is not responsible for the contents or reliability of any linked Web sites and does not necessarily endorse the products, services, or information described or offered within them. We cannot guarantee that these links will work all of the time and we have no control over the availability of the linked pages. Trademarks Avaya and Communication Manager are trademarks of Avaya Inc. equality is a registered trademark, and Quality is a trademark of Verint Systems, Inc. All other trademarks identified by the or are registered trademarks or trademarks of their respective owners. Third-party software license agreements This computer program is protected by U.S. and international copyright laws, patent laws, and other intellectual property laws and treaties. Unauthorized use, duplication, publication and distribution of all or any portion of this computer program are expressly prohibited and will be prosecuted to the maximum extent provided by law. Your rights in this computer program are limited to the license rights granted under the license agreement executed by you in hardcopy form (or if none, by acceptance of the clickwrap terms included with this computer program). If needed, please contact your vendor for an additional copy of those terms. All other rights, title and interest are expressly restricted and retained by Verint Systems, Inc. and its licensors. The following open source applications are included with this computer program ("Open Source"), with ownership in those applications as indicated below. Ant, Axis, log4j and Tomcat are each copyright The Apache Software Foundation. All rights reserved. Licensed under the Apache License, Version 2.0, which can be obtained at PostgreSQL is portions copyright , The PostgreSQL Global Development Group, and portions copyright 1994, The Regents of the University of California. All rights reserved. A copy of the PostgreSQL license can be obtained at jcifs is copyright 2004 The JCIFS Project. All rights reserved. Licensed under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, which can be obtained at JoeSNMP is copyright Blast Internet Services, Inc, and copyright Oculan Corp. Copyrights for modified and included code are included in the individual source files. All rights reserved. Licensed under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, which can be obtained at A license in each Open Source software application is provided to you in accordance with the specific license terms specified above. EXCEPT WITH REGARD TO ANY WARRANTIES OR OTHER RIGHTS AND OBLIGATIONS EXPRESSLY PROVIDED DIRECTLY TO YOU FROM VERINT, ALL OPEN SOURCE SOFTWARE IS PROVIDED "AS IS'' AND ANY EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE OWNERS OF THE OPEN SOURCE SOFTWARE OR ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THE OPEN SOURCE SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. Avaya support Avaya provides a telephone number for you to use to report problems or to ask questions about your contact center. The support telephone number is in the United States. For additional support telephone numbers, see the Avaya Web site: Select Support, and then select Escalation Lists. This Web site includes telephone numbers for escalation within the United States. For escalation telephone numbers outside the United States, select Global Escalation List.

3 Verint ContactStore for Communication Manager Release 7.8 Planning, Installation and Administration Guide About This Guide Intended audience Summary of information included in this guide Conventions used in this guide Additional references Introduction What s New in Version Server components Verint ContactStore for Communication Manager Server Optional Server Applications End-User tools Quality Monitoring Bulk Search and Replay Administration Tools Recording Functionality Sampled Recording for Quality Assessment Bulk Recording Modes Ad-hoc or Occasional Recording Modes Replay Options Live Monitoring Recording Methods Service Observe Single-step Conference Conference Miscellaneous Beep Tone International support Liability Introduction Audio Format G Issue 5 May,

4 Contents G.729A Storage Requirements Storage at Each Recorder Quality Monitoring Central Database Storage Archive Call Storage Backup Storage Server Platform Sizing DVD+RW Drive Network Issues Load Ports Used Network Address Translation Routing Recorder Licensing Server Licenses Port Licenses Other Licenses Timed Trials Avaya system prerequisites Communication Manager Gateway Resources AE Services Expansion Interface Boards (TN570) C-LAN VoIP Resources Multi-Connect Capacity DMCC (IP_API_A) Licenses TSAPI Simultaneous User Licenses VoIP Network Design Topologies Bulk Recording System Sampled Quality Monitoring (only) System Bulk Recording + Quality Monitoring System Large Bulk Recording Systems Integrating with other systems Supplementary Tagging of Bulk Recordings Explicit External Control of Bulk Recording Enhanced Quality Monitoring Overview Avaya System Configuration Prerequisites Communication Manager Configuration AE Server Configuration TSAPI Configuration Test Phonesets Order in which to Install Applications Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

5 Contents Platform Prerequisites Operating System DVD Drive Network Connectivity Installing Verint ContactStore for Communication Manager Installing Verint equality Balance (V5) Installing Verint equality Balance (V7) If using selective quality recording If using "ContactStore Plus" style quality recording Installing Viewer Prerequisites Installing Viewer Create an MMC Console Configuring Viewer Configuring Verint ContactStore for Communication Manager Installing Archive Overview Accessing the System URL Initial User Account Key Points Licensing Terminology Obtaining a License Activation Key Adding additional licenses Reinstalling on the same PC Reinstalling the Recorder on a new PC Security Securing the System Windows Authentication System Settings Server Communication Manager System Monitoring Port Allocation Common Settings Assigning Ports On Demand Recording Meeting Recording Station Bulk Recording Station Executive Recording Conferenced Recording Quality Server Ports (for Selective Quality Recording) Unify or Externally Controlled Ports Phone Replay Ports Live Monitor Ports Issue 5 May,

6 Contents Search and Replay Search and Replay Access Rights ActiveX Control Download Modify Default Behavior Backup/Restore Application Backing up the Database Restoring data to a new PostgreSQL database Backing up Voice Recordings Distributing User Instructions Those Using Recording or Live Monitor Those entitled to replay calls Configuring Avaya Support Remote Access Introduction Status Monitoring Alarms and Events System Overview Port States Peak Activity Audit Trail Local Archive DVD+RW media control and care Loading a disk Changing disks Labeling a disk Preventative Maintenance Daily Weekly Monthly Every Six Months Use of SSL Windows Domain Authentication Blocking Replay from the Recorder Changing Passwords Linux User Accounts Postgres Database Owner Recorder Account on eware/viewer SQL Server Encrypted File Storage PCI Compliance Properties File Standby Server Central Replay Server Installation Configuration User Defined Fields Storing UDFs Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

7 Contents Search and Display UDFs UDF Support in Viewer Viewer Compatibility Customizing Search and Replay Default Configuration Files Custom Layout Files Internationalization Usage Report Enabling the Report Content Accessing through URL: Accessing the Usage report in a log file Selective Record Barring Configuration Example Limitations Recording files WAV files XML files Internal Database Recording details Configuration details Recorder Interfaces HTTP Interfaces Offered Communication Manager equality Balance V5 or V7 in Selective Recording Mode equality Balance V7 in "ContactStore Plus" mode Other Recorders External Control Interface Database Upload Interface Summary Hints and Tips Where to Look for Clues Determining Current Version Application Logs Tomcat Logs Remote Access Specific Problems System Administration page problems Connectivity Search and Replay problems Recording Problems Live Monitor Alarms Alarms Table Introduction When to Use External Control Issue 5 May,

8 Contents This Appendix Port Allocations Master + Slave Systems Java API Toolkit TCP/IP Protocol Overview Connection Method General Protocol Specification Channel Identification XML Tagging Basic Call Tagging Fallback Mode Examples Third-party CTI Control Additional Call Tagging TCP/IP Message Sequences Redundant SAN Duplicated recording Standby Recorder Options Prerequisites for high availability Standby recorder licensing Automated Configuration Shadowing Manual Configuration of Standby Known limitations Distributed Systems Supported failure modes Standby recorders and Unify/External Control Supported Master/Standby Topologies Mode of operation Standby configuration (automatic) Standby configuration (manual) Power-On Standby mode Failure Detection Disk Space Monitoring Active mode Return to Standby mode Switchover Implications Restoring the Primary Comparison with hardware switch-over units Standby Recorder Configuration Configuration Differences Upgrade Process for Primary/Standby Overview Modes of Operation Principles of Operation High Availability Installation Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

9 Contents Configuration Programmer s Guide Enable(in string station, in SeqCouple tags) Disable(in string station) Start(in string station) Stop(in string station) Tag(in string station, in string inum, SeqCouple tags) Assign() Overview Principles of Operation Installation Configuration Choosing the dialler version Changing the log level Controlling the service Automatic restart Overview Disks NICS DVD Soundcard Kickstart Installing Unlimited Strength Encryption Installing a Signed SSL Certificate Selecting a Certificate Authority (CA) Backing up the Keystore file Creating the new Certificate Generating a Certificate Signing Request Importing the CA's certificates Backing up the keystore file Changing Tomcat Port Numbers Encrypting Properties File entries Configuring Viewer and Archive to use HTTPS Importing the public certificates into Viewer and Archive Configure Viewer to use https and a secure port Configure Archive to use https Configuring Archive for a Key Management Server Configuring Archive Signing and Delivering the certificate Installing the Certificate Issue 5 May,

10 Contents 10 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

11 About This Guide The Verint ContactStore for Communication Manager Planning, Installation and Administration Guide provides details of the Verint ContactStore for Communication Manager system, as well as recommended and required components. Issue 5 May,

12 About This Guide Intended audience This guide is designed for use by: I.T. managers Installers System Administrators Summary of information included in this guide The following table provides information about this guide. Chapter Title Chapter 1: System Overview Chapter 2: Planning and Prerequisites Chapter 3: Installation Chapter 4: Configuration Chapter 5: Operation, Administration and Maintenance Chapter 6: System Security Chapter 7: Advanced Configuration Appendix A: Technical Reference Glossary Description This chapter provides an overview of the design options for Verint ContactStore for Communication Manager. This chapter gives details of the prerequisites for Verint ContactStore for Communication Manager. This chapter gives details of the steps to install Verint ContactStore for Communication Manager. This chapter gives details of the steps to configure Verint ContactStore for Communication Manager. This chapter provides detais of regular maintenance required by Verint ContactStore for Communication Manager. This chapter sets out the available and recommended steps you can take to secure the system. This chapter details the rarely used advanced settings of Verint ContactStore for Communication Manager. This appendix provides technical details about Verint ContactStore for Communication Manager. The glossary defines the terms you need to understand Verint ContactStore for Communication Manager. 12 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

13 Conventions used in this guide Conventions used in this guide The following table shows how user input, output and instructions are highlighted in this guide, as well as special notations that you will see as you use this guide. To show... This style is used For example... Information shown on screen Characters that you should type exactly as shown Characters that you should replace with appropriate information Menu selections, buttons and tabs Helpful hints that can improve the efficiency or effectiveness of your work Important details that we want to make sure that you do not overlook Advice that can help you avoid undesirable results Fixed width Fixed width, bold Fixed width, bold italic Sans Serif, Bold Tip: Note: You should see the prompt below: login: Enter the following command: mount /mnt/cdrom Browse to the new server by entering Click on the Install button. Tip: If no part-time licenses are available, a full time license may be used instead. Note: Media Encryption may or may not show up on this form.! Important:! Important: If the network does not meet the three conditions listed, there will be no media resources. Issue 5 May,

14 About This Guide To show... This style is used For example... Situations that can result in: Harm to software Loss of data An interruption in service Situations that can result in harm to hardware or equipment! CAUTION:! CAUTION:! WARNING:! WARNING: Perform this procedure only after normal business hours. This procedure restarts all links on the interface, and can cause a temporary loss of service. Make sure that the disks are the Update you require. RedHat and other vendors still sometimes supply Update 0 disks. Additional references The following guides contain additional information you may find helpful. Avaya Communication Manager Call Recording: A Design Approach for Device Media and Call Control (DMCC, previously called CMAPI) (Compas ID ) Verint ContactStore for Communication Manager User Guide Viewer for Communication Manager Installation Guide Viewer for Communication Manager Quick Reference Guide Contact Archive for Communication Manager Installation Guide Contact Archive for Communication Manager System Administration Guide Unify for Communication Manager Installation Guide Witness Enterprise Deployment Guides Impact 360, Full-time Recorder, Enterprise Security Administration Guide Release 7.8 Avaya Communication Manager Guide to ACD Contact Centers Administrator s Guide for Avaya Communication Manager Administration for Network Connectivity for Avaya Communication Manager Note: Note: Avaya Communication Manager documentation is available through the Avaya online support Web site, 14 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

15 Chapter 1: System Overview This chapter provides an overview of the design options for a Verint ContactStore for Communication Manager system. Issue 5 May,

16 System Overview Introduction Verint ContactStore for Communication Manager provides an extremely efficient and scalable, software only, voice recording platform, running on standard PC Hardware. It uses Avaya's Device, Media and Call Control (DMCC) features to provide a wide range of recording modes with all the benefits of VoIP-based recording but without the limitations of passive tap IP recording systems. This new approach to recording offers the following benefits: The recorder can record potentially any call on the switch. Traditional trunk and extension modes cannot record internal and tandem calls respectively. There is no cabling to maintain as new trunks or extensions are added to the switch. Uses standard PC servers with no proprietary cards. This chapter describes the major components of the Verint ContactStore for Communication Manager system. It describes the Verint ContactStore for Communication Manager (which is a mandatory component) and the optional components 16 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

17 Introduction What s New in Version 7.8 Version 7.8: Provides a range of enhanced security options as described in Chapter 6 Supports Redhat Enterprise Linux (RHEL) Versions 4 and 5. (Existing installations on RHEL 3 can be upgraded to CSCM 7.8 but this version is no longer supported for new installations). Supports Avaya Communication Manager Version 4 Supports numbering plans up to 16 digits Supports No-talk Service Observe - which does not require a time-slot on your port network. Allows you to update call tagging details after recordings have completed. Issue 5 May,

18 System Overview Server components The Verint ContactStore for Communication Manager system can be installed as a single server solution providing recording and replay of calls. Large systems can be built from several independent recorders or in a master/slave topology. The former approach works well for station-side recording while the latter provides many of the benefits of traditional trunk-side approaches - without most of the disadvantages. Internal calls can still be recorded, for example. You can extend the scope of the system by adding additional optional server applications in order to: perform quality monitoring and screen recording provide centralized shared search and replay provide centralized archive facilities These optional server applications require their own physical server. All of the other applications in the suite require Windows servers. Note: Note: This Guide is the detailed reference for the Verint ContactStore for Communication Manager (CSCM) only. This component can record voice calls over IP using Avaya DMCC softphones. It can also act as the Master - controlling other recorders that provide additional capacity. Each of the other components has a corresponding guide (as detailed in Additional references on page 14), which you should refer to for more detail. Verint ContactStore for Communication Manager Server In most small to medium-sized (up to a few hundred channels) Bulk Recording systems, one of these applications provides the entire recording and replay system (as shown in Introduction on page 16). In larger systems (where a single physical server is not powerful enough), you should install multiple instances of this application on different physical servers, each providing a subset of the system's overall functionality. The Verint ContactStore for Communication Manager can: connect to the Avaya Communication Manager Server via Application Enablement Services (AES) and control all voice recordings record and store telephone calls via IP archive the recordings it makes to a local DVD+RW drive 18 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

19 Server components provide basic search and replay services to users connecting via their browser or via their telephone provide voice recording services to the Verint equality Balance (V5 or V7) application control other Verint ContactStore for Communication Managers - known as Slaves be controlled by another Verint ContactStore for Communication Manager - i.e. act as a Slave act as a warm Standby to a Verint ContactStore for Communication Manager Master act as a dedicated, centralized replay server, holding details of recordings made by other Verint ContactStore for Communication Managers Optional Server Applications Verint equality Balance Application Viewer This application provides the features needed to automate your agent quality assessment process, including: Design of evaluation forms Various call selection methods Screen content recording and storage Performance recording and reporting See Verint equality Balance Administration, Evaluations, Graphic and Reporting, Server Infrastructure, Troubleshooting and Installation Guides for full details of this component and Installing Verint equality Balance (V5) on page 88 for how to integrate it into the overall system. Although the Verint ContactStore for Communication Manager application includes search and replay facilities within it, you will need to install this alternate replay server in the following cases: If you wish to deploy centralized Archive If you wish to combine recordings with those made by other Verint recorder types In such cases, you may install Viewer on an additional server and configure each of your recorders to copy the details of its recordings to the database on this server. The database holds detailed information on all recordings including where the audio content is stored (for example, either on individual recorders or archived offline). The following figure shows a Issue 5 May,

20 System Overview typical system, using Viewer to provide access to recordings from two independent recorders. Archive See Viewer Installation Guide for full details of this component and Installing Viewer on page 90 for how to integrate this application into the overall system. Each recorder can archive the recordings that it makes itself via a single, locally connected DVD+RW drive. If you need more sophisticated archival - from multiple recorders to a common destination, such as a Network Attached Storage (NAS) device or of selected 20 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

21 Server components recordings only, you can deploy the Archive application on a separate server as shown below. External Control See Archive Installation Guide for full details of this component and Installing Archive on page 94 for how to integrate this application into the overall system. Note that you must first deploy Viewer in order to use Archive. In addition to the data provided by the AES links, you may wish to control and/or tag your recordings with details from other CTI feeds or application interfaces. These may include third party systems and/or your own in-house applications. The recorder supports a wide range of systems and allows them to be connected to your recording system. These include: Avaya Interaction Center (AIC) Avaya Proactive Contact (formerly Predictive Dialing System, PDS) Additional Verint Systems applications As an integral part of Verint Systems' range of recording and analysis products, the following applications can be layered on top of the recording system: Issue 5 May,

22 System Overview equality Now (agent training tools) equality Vision (advanced recording visualization) 22 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

23 End-User tools End-User tools To access the recordings held in the system, users have a variety of options. Quality Monitoring Supervisors and quality assessors use a Windows application (V5) or browser (V7) to evaluate recordings - using both voice and screen content where available. They can assess and review calls with their staff as well as analyze and report on the results of assessments. Audio is replayed via the desktop PC's soundcard or via the telephone. Live monitoring of audio and screen content is also supported. Verint equality Balance Evaluations Guide provides full details of the Quality Monitoring application. Bulk Search and Replay Integral Search and Replay The Verint ContactStore for Communication Manager includes a search and replay application within it. This replay mechanism is a very simple and intuitive browser-based interface, requiring the user to access it via Internet Explorer Version 5.0 or higher. The Search and Replay application is hosted on a web server running on the recorder itself. It uses a local database of recordings to allow users to search for bulk and ad hoc recordings by: Call start date/time The name(s) and number(s) where provided of any party on the call (including ANI and/or DID where provided by the switch) Agent ID and name Call duration Universal Call Identifier (UCID) Service Name and Number (also known as Vector Directory Number, VDN) User defined fields supplied by external controllers Issue 5 May,

24 System Overview! Important: Important: Verint ContactStore for Communication Manager does not retain those calls recorded for Verint equality Balance; these use that product's own replay mechanism. Access restrictions determine which calls individual users are able to replay. Each recording is assigned one or more "owners" at recording time (see Search and Replay Access Rights on page 136 for further details). The user can play and view details of any call that matches their search criteria and access rights. When a call is played, a graphical representation of the audio level of the call, the audio wave form, is displayed. The audio wave form shows silence and tones, so the user can click beyond irrelevant sections and pinpoint parts of the call that are of interest. See the accompanying User Guide for further details of this application. Central Search and Replay Where multiple Verint ContactStore for Communication Managers are deployed, an additional server can be nominated as a Central Replay Server. This server is not used for recording but can provide telephony replay ports. The other recorders upload details of the recordings they have made into this server's database allowing users to search for and replay recordings made on any recorder without having to know which one recorded a particular call. External Search and Replay The recorder also supports replay from other applications including equality Balance (V5 and V7), Contact Viewer and Contact Vision. Administration Tools As the suite is designed specifically for Avaya systems, much of the complexity associated with generic recording systems has been removed resulting in a system that is easy to configure and maintain. With the exception of the Quality Monitoring (V5) system (which is a Windows application), the recorders are administered via a web interface. The detailed use of this interface is the subject of later Sections in this guide. 24 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

25 Recording Functionality Recording Functionality The Verint ContactStore for Communication Manager system records telephone calls made on Avaya Communication Manager release 3.0 and above - using analog, digital or IP-based stations. The recorder hosts a bank of IP softphones - each of which provides a recording or replay port. These record calls by conferencing into a live call using either the Service Observe or single-step conferencing features of the switch. This manual only considers audio recording load. For Screen Content recording, please refer to the Verint equality Balance (V5 or V7) Server Infrastructure Guide. The first task in designing any recording system is to define what is to be recorded. This in turn is often driven by the reason for wanting the recordings. As the recorder's ports can also be used for replay and live monitor, it is important to consider these together before specifying a system. This section introduces the various ways in which recorders' ports can be used but you should refer to Port Allocation on page 114 for detailed functionality and limitations of each mode. Sampled Recording for Quality Assessment If the end goal is to record only a sample of calls in order to assess the quality of your interactions with customers, Verint equality Balance (V5 or V7) can be used alongside the Verint ContactStore for Communication Manager. See this product's Administration, Evaluations, Graphic and Reporting, Server Infrastructure, Troubleshooting and Installation Guides to determine how you will use the Quality Monitoring system to control recording. Quality Recording Channels (equality Balance V5 only) From the above manuals, you must determine the number of concurrent audio recording "channels" that will be required on the Verint ContactStore for Communication Manager to support your Quality Monitoring requirements. Quality Replay Channels Note that you will also need to consider how the Quality Application is to replay calls for supervisors to review and coach their staff. If you choose to replay via their telephones, then you must provide sufficient ports for this. This is in addition to the recording channels. The overall port count required for selective quality recording is therefore the sum of recording and replay requirements. Issue 5 May,

26 System Overview Licensing The Verint ContactStore for Communication Manager must be provided with a license key that includes the required number of Quality Channels (the sum of recording and replay ports assigned to the Quality Monitoring application). Bulk Recording Modes If you need to record all of the calls taken by specific stations, agents, skill groups or Vector Directory Numbers (VDNs) then you require Bulk Recording - of which there are three variants. Station Bulk Recording This recording mode can be used to record all calls occurring at specific stations. Advanced configuration options let you filter the calls by VDN or Skill Group rather than have to record every call on the nominated stations. This mode uses the switch s service observe feature and therefore requires a dedicated port per station that is recorded. Refer to Station Bulk Recording on page 121 for a full description of this mode and how to configure it. Conferenced Recording Rather than dedicating ports to specific stations, a pool of ports can be used in conjunction with an Avaya CT (also known as TSAPI) link to record calls on specific stations, agents, skill groups or VDNs. In this mode, the recorder uses single-step conferencing to connect into the calls to be recorded. Refer to Conferenced Recording on page 125 for a full description of this mode and how to configure it. Externally Controlled Recording More complex recording requirements may be met by customized or specialist applications that interface to other CTI feeds or customers' own applications. Such an application can control ports on the Verint ContactStore for Communication Manager, allowing it to record exactly what and when it requires. The application may also "tag" the recordings with additional details such as customer number or account number. Refer to Unify or Externally Controlled Ports on page 131 for a full description of this mode and how to configure it. 26 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

27 Recording Functionality Which to use? In essence, Station Bulk recording is simpler and more economic to deploy than Conferenced mode. Against this, however, it is a little less flexible. Externally controlled recording is normally only considered when complex recording rules or additional CTI control is needed. It is imperative that you review the detailed functionality, limitations and caveats in Station Bulk Recording on page 121 and Conferenced Recording on page 125. Bulk plus Quality Licensing Where Quality assessment is to be performed as well as bulk recording, Verint ContactStore for Communication Manager can be used alongside Verint equality Balance (V7.6). The recorder advises the quality system of all recordings that it makes and these can be used for quality assessment. A "full-time recording channel license" is required for each recording port assigned to these modes. In Station Bulk, this equates directly to one such channel license per station to be recorded. In conferenced mode, a channel license is required per port assigned to this mode. Ad-hoc or Occasional Recording Modes Three recording modes are provided for those requiring occasional recording: On Demand Recording This mode lets users of any phone on your system dial into or conference in a recording port as and when they want to start recording a call. One or more "pools" of ports on the recorder can be assigned to this recording mode and accessed via Hunt Group numbers so that callers automatically reach an available port. The recorder automatically answers the incoming call on its port and starts recording. Refer to On Demand Recording on page 119 for a full description of this mode and how to configure it. Meeting Recording A novel use for recording is in taking a detailed log of a meeting, either as an audio record for those attending, or as a way to include non-attendees later. You can use any meeting Issue 5 May,

28 System Overview room or office with a telephone that has a speakerphone or, ideally, conference phone capabilities to record the meeting.! Important: Important: The audio recorded with Meeting Recording is the same as someone dialing in would hear it on the phone used to record it. Place the phone so it picks up the speech of all participants. They should speak loudly and clearly. Experiment with this recording mode before relying on it to provide full and complete records of your meetings. Verint cannot be held responsible for the failure to pick up all of the audio intelligibly. Use this recording mode as an aid to note taking, not a replacement for it. One or more "pools" of ports on the recorder can be assigned to this recording mode and accessed via Hunt Group numbers so that callers automatically reach an available port. The user follows the spoken instructions to start the recording and specify which user(s) can access it. Refer to Meeting Recording on page 120 for a full description of this mode and how to configure it. Station Executive Recording Scope This option lets users of specific phones choose which of their calls are recorded. Unlike the previous two modes, this method records the call from the beginning. Unless the user chooses to keep the recording it is deleted as the call ends (or shortly afterwards according to how the system is configured). This recording mode is particularly suited to users who occasionally need to record a call, and who might only recognize the need to record after the call is in progress or after it has ended. The classic example of this is in "bomb-threat" or "abusive call" recording. In such cases, it is only once the nature of the call becomes apparent that the agent realizes that he should record the call. The advantage of this recording mode over other recording techniques is that the entire call is being recorded all the time. The whole call can be retained even if the nature of the call and the need to retain it is not recognized until the end of the call or, if the delayed deletion option is enabled, after the call ends. Refer to Station Executive Recording on page 123 for a full description of this mode and how to configure it. Ports assigned to On Demand or Meeting recording can be assigned to one or more hunt groups making them easily shared across one or more user populations can be used not only by stations on the switch but also from outside the switch if they are made accessible via a DID number. 28 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

29 Recording Functionality Licensing Station Executive recording ports on the other hand are dedicated to specific stations and can only be used to record those stations that can be service observed. As On Demand and Meeting recording ports are shared by all users, they typically record many calls per day and hence a "full-time recording channel license" is required for each port assigned to these modes. Although Station Executive recording requires a dedicated port on the recorder for each station that uses it, Verint recognizes that typically only a few recordings are made per station per day. Hence the value of such a recording port is less - and a "part-time recording channel license" is used. Tip: Tip: If no part-time licenses are available, a full time license may be used instead. Replay Options Soundcard Replay Many users choose to replay recordings via their browser and the soundcard on their PC. This does not use any ports on the recorder and does not require any additional licensing. See the equality Balance (V5) manuals for details of how to use this replay mechanism in conjunction with this application. Telephone Replay However, the recorder also supports replay via the user's telephone and this does use a port on the recorder. equality Balance (V5) replay ports form part of the overall pool of ports assigned to Quality Recording. All other replay applications make use of telephony Replay ports on a Verint ContactStore for Communication Manager. Ports configured in this way are compatible with Verint "AudioServer" ports. Refer to Phone Replay Ports on page 133 for a full description of this mode and how to configure it. Issue 5 May,

30 System Overview Licensing Replay channels that form part of the equality Balance (V5) port pool require a Quality channel license. Telephony replay ports used by all other applications require a replay channel license. Live Monitoring Via equality Balance (V5) Users of equality Balance (V5) can listen in to calls in real-time using their telephone. This uses one of the replay ports within the Quality pool. See the equality Balance (V5) manuals for details of this live monitoring mechanism. Live Monitor Ports Licensing On Communication Manager 3.x, a side-effect of using service observe to record calls (as used by Station Bulk and Station Executive recording modes) is that this switch feature is no longer accessible to supervisors who wish to use it. To mitigate this, one or more ports on a Verint ContactStore for Communication Manager can be assigned for Live Monitor use. Supervisors can dial into these ports and dial the number of the station they wish to observe. The audio being recorded is relayed to them giving similar functionality to service observe. With Communication Manager 4.0 and above, two service observe sessions can be in progress on the same call and it is therefore unlikely that you will need to use Live Monitor ports. Refer to Live Monitor Ports on page 133 for a full description of this mode and how to configure it. Live monitoring via equality Balance (V5) uses a replay port within the Quality pool and is therefore controlled by the overall Quality Channels license count. Live Monitor ports on the recorder are licensed separately and a Live Monitor channel license is required for each one. 30 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

31 Recording Methods Recording Methods The recorder uses three different methods to record calls. The table below shows which mode is used by each mode. Recording Mode On Demand Recording Meeting Recording Station Bulk Recording Station Executive Recording Quality Monitoring for equality Balance V5 Externally Controlled (Service Observe mode) Conferenced Recording Externally Controlled (Single-step conference mode) where the external controller instructs the recorder to establish a single-step conference. Externally Controlled (Single-step conference mode) where the external controller is responsible for connecting the recorder port into the call to be recorded. Uses Conference Service Observe Single-step Conference Depends how the external controller establishes the call. May be conference or single-step conference. The recording mechanisms differ in several respects as described below. Refer to Port Allocation on page 114 onwards for details of the pros and cons of these modes as these must be considered alongside the issues listed here. Note: Note: Regardless of which recording method is used, each recorder port that joins a call to record it counts as an additional party on that call. Hence your normal limit of 6 parties on a call includes all of the recordings being made. Every recording made reduces the number of "real" participants you can have on the call. Issue 5 May,

32 System Overview Service Observe Station Bulk, Station Executive and Quality recording modes use the Communication Manager's Service Observe feature. This has the following characteristics. Limited Number of Service Observers per Station In Avaya Communication Manager 3.x, only one other station can service observe a station at any one time. Therefore: a supervisor cannot use Service Observe to monitor stations or calls that are being recorded. They will either fail to observe the call or will block the recorder from recording it. To listen in to calls (in real time) on stations that are being recorded, users must dial into Live Monitor ports on the recorder or use equality Balance (V5) s Live Monitor feature. Live Monitor ports can be included in any license. If you feel that Live Monitor ports may be useful, request them when ordering the system. Quality Monitoring uses service observe. If a station is being recorded in another mode using service observe, then Quality Monitoring can only work if the quality and other port recording it are on the same recorder. If you need to perform bulk and quality recording, use a single recorder or, if the load is too much for a single server, you must use a single-step conference mode for the bulk recording. In Avaya Communication Manager 4.x, up to two stations can service observe a given station. It is therefore rare that Live Monitor ports are needed in this case. However, it is still a good idea to record a station on a single recorder where possible - due to the per call limitations described below. Limited Number of Service Observers per Call In Avaya Communication Manager 3.x, if more than one party on a call is being service observed, then only one of the observers will actually receive the audio from the call. The other will be completely unaware that a call is in progress. Therefore: A call between a Station Bulk recorded station and a Station Executive recorded station can only be visible to one or other of the two recording ports. Which one depends on which station made the call. If the station executive port initiated the call, it will be deleted unless the user deliberately retained it. An attempt to Quality Monitor a station on an internal call will fail if the other station on the call is the one receiving the audio. If you need to quality monitor agents who solely or predominantly take internal calls and are using bulk recording as well, you should use a single-step conference mode in preference. If a user service observes a VDN, they will stop any call they hear from being recorded. Use Class of Restriction (COR) settings to prohibit users from observing VDNs that include calls that you are trying to bulk record. 32 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

33 Recording Methods In Avaya Communication Manager 4.x, each call can be service observed up to twice. Normal, two-way internal calls can therefore be recorded twice and it is only with three (or more) way calls that the above limitations apply. It is still important to avoid service observing VDNs related to calls that you need to record. Assign Ports to Stations not Agents/VDNs Although Communication Manager can service observe VDNs and Agents, the recorder only supports service observe of station numbers for the following reasons: Call segmentation Service observing a VDN only results in one call at a time on that VDN being observed. This is not appropriate for a bulk recording mode. Service observing an agent id results in failure if that agent is not logged in. The recorder cannot distinguish between failure due to this reason and other failure modes. It would therefore raise alarms continuously. As the recorder port sees the same activity as a real user would do when service observing a station, audio ceases to flow when the station being observed places a call on hold. Placing a call on hold (automatically as part of a transfer/conference setup or deliberately) therefore stops recording. A new recording is started when the call is retrieved from hold. The segments of the call are stored as separate recordings (.wav files) and can be searched and played as two calls.! Important: Important: If the call is retrieved from hold by another station, either by means of a transfer or a "park" operation, the subsequent segment of the call is only recorded if the other station is also being recorded. The recorder's Search and Replay mechanism treats these segments as two separate calls. However, if you find one call segment, you can easily find the other segments by clicking on the link in the UCID column. Viewer, a separate server replay application, on the other hand, "merges" these call records into one and shows the hold period as silence. If an external controller is used, it may segment calls at points it chooses. Bulk Recording requires dedicated port Any attempt to service observe a station will - on perhaps one in a thousand attempts - fail. While this is not a significant issue for a sampling quality recording system, it is unacceptable for bulk recording. This is why the recorder must dedicate a port to each phone that it bulk records via service observe. By doing so, it can establish service observe as the recorder starts - and retry on the rare occasion that this fails - and then hold the service observe permanently. This provides reliable recording. Issue 5 May,

34 System Overview Personal Station Access (PSA) PSA is not supported for the same reasons that agents cannot be targeted. As the address is not always present, the recorder cannot maintain a service observe on it at all times as is required for bulk recording. Internal Calls via a VDN do not show who answered - CM 3.1 and earlier If an agent handling a call dials a VDN to transfer, conference or simply consult on it, the details of the party that answers the call are not visible to a station service observing that call. The call will be tagged with the VDN number dialled but not the agent or station that answered the call. This is corrected in CM3.1 load 632. If using an earlier version than this, consider updating with Patch # Transfer via Coverage Bridged Lines Exclusion Beep Tone Calls transferred to a group with coverage (e.g. Coverage Answer Group) and not picked up by the original destination address cannot be observed. To overcome this limitation, use a phantom extension and transfer the call to that rather than directly to a group with coverage. Where bridged line appearances are used on stations being recorded, calls made on the secondary line appearance will be recorded but the details about the call will be limited to the identification of the station being recorded. Only calls made on the primary line appearance will include full call details. If any party on the call invokes the Exclusion feature, the call cannot be recorded. (Check with Avaya for changes to this from CM 3.1 onwards). If you do not need to inject beep tone while recording, use "No Talk" service observe. This option, available from CM 3.1 onwards, does not use up a time-slot. Otherwise you must design for an additional time-slot being used to carry each call that is actually being recorded. If you need to inject beep tone you must use "Talk/Listen" service observe. "Listen Only" service observe should not be used with CSCM. 34 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

35 Recording Methods Single-step Conference CTI Requirement Timeslots Single-step conferencing as used by Conferenced mode recording and Externally controlled (single-step conference option) has the following issues: When recording in this mode, the recorder uses JTAPI which in turn uses Avaya CT (also known as TSAPI). This requires not only that the CTI feed be available but also that sufficient licenses are available for it. See TSAPI Simultaneous User Licenses on page 60 for details. Single-step conferences do not require an additional timeslot on the switch if beep tone is NOT used. Call Segmentation The recorder continues to record only so long as the real parties on the call are connected. If the call is on hold, recording stops. If an external controller is used, it may segment calls at points it chooses. Bridged Lines not Supported Beep Tone Because of an inherent limitation in the underlying call/connection model, conferenced recording does not support bridged line appearances. You must not attempt to record bridged lines in this mode. If you choose to inject beep tone on a single-step conferenced recording, the recorder becomes a full member of the call rather than a listen-only member and therefore: Becomes visible to the agent, who can see that the call is conferenced Uses an extra timeslot on the switch Conference In some modes a user or an external application will dial a port on the recorder. When this happens, the recorder answers the call and is therefore a normal party on the call. Issue 5 May,

36 System Overview Timeslots As "just another party" on the call, the recorder port will use the single additional timeslot that any other phone would use when added to a call. Call Segmentation As the recorder port is a normal party on the call, it is still connected even if one or more other parties on the call places the call on hold. It will receive the same audio that the parties remaining on the call receive. This may include music on hold or silence. If an external controller is used, it may segment calls at points it chooses. 36 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

37 Miscellaneous Miscellaneous Beep Tone Some states and countries require that both parties on a call be made aware that the call is being recorded. One way to do this is to apply a tone to the line. Note, however, that most telephone users are unaware of the significance of this tone and might, in fact, regard it as a fault. There are more effective means of informing the user that the call is being recorded. For example, you can inform users on advertising and in contract literature or play a recorded announcement before the call is connected. Check the legal position in your jurisdiction and apply the appropriate settings. See the discussions of how each recording mode works earlier in this chapter to determine whether or not you need to turn on this warning tone within the recorder. In most cases, it is recommended that you set this option to No and use the other mechanisms described there. International support The recorder's browser-based Administration and Search/Replay interfaces are provided in several languages. Check with Verint for availability of specific languages. International support includes: Time zone and DST support. All dates and times are stored in the database in Coordinated Universal Time (UTC). However, when you view records using the search and replay application, these are converted to your local time. If you view the records using a database query tool, the times will be shown in the time zone of the client machine, which may be different from the server time. Note that the XML files relating to the recordings include ISO standard timestamps, giving both UTC and offset from Greenwich Mean Time (GMT). Liability Liability of Verint for failure to record any calls is limited under the terms of supply. Issue 5 May,

38 System Overview 38 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

39 Chapter 2: Planning and Prerequisites This chapter gives details of the prerequisites for a Verint ContactStore for Communication Manager system. You should also review Chapter 6 System Security as some of the optional elements described there may also require additional cost and/or effort. The main sections in this chapter are: Introduction on page 40 Audio Format on page 41 Storage Requirements on page 43 Server Platform on page 46 Network Issues on page 51 Recorder Licensing on page 52 Avaya system prerequisites on page 54 Topologies on page 62 Integrating with other systems on page 67 Issue 5 May,

40 Planning and Prerequisites Introduction Unfortunately, there is no one "right" order in which to plan a system. A number of the requirements and pre-requisites are inter-dependent and it may be necessary to iterate through this section several times, refining your plans each time. This Chapter assumes that you have read the previous one and know the type and quantity of recording, replay and live monitor that you require - and hence your total license requirements. This Chapter will now guide you through: identifying what is to be recorded, the options for doing so and the relevant license requirements for the preferred option determining the storage needed to hold your recordings sizing the servers needed for each application quantifying the network bandwidth needed You can then determine an appropriate system topology. 40 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

41 Audio Format Audio Format In all cases, calls are recorded using resources on the Avaya Communication Manager to conference a recorder port into a live telephone call. One of the benefits of this approach is that the recorder can ask to receive audio in a format that suits it - without impacting the experience of the parties actually speaking on the call. The recorder can request either G.711 (µ-law), or compressed G.729A data streams. You must choose which is appropriate for your recording needs, network and server sizing. G.711 If the recorder receives audio in G.711 (µ-law) rather than G.729A, bulk recordings will be normally be compressed by the recorder and stored as G.726 (16kbps) files. (Compression can be disabled by setting cscm.disablecompress=true in the properties file.) Selective Quality recordings controlled by equality Balance will be passed to it unchanged. Note: Note: When using G.711, the recorder always requests µ-law, never A-law. This applies to all countries regardless of that country s national preference. Therefore, if you select G.711 rather than G.729A the implications are: For All Recording Several times more bandwidth will be needed between the media processing resources and the recorder. Less Avaya media processing resource will be needed. For Bulk Recording The capacity of a given server will be about halved The quality of recordings will be a little lower The storage space used will be doubled For Quality Recording The quality of recordings will be a little higher The storage space used will be eight times higher (This is for voice recordings. Note that when combined with screen recordings, the overall impact is proportionately less). Issue 5 May,

42 Planning and Prerequisites G.729A Conversely, if receiving G.729A, the recorder will not have to perform any compression tasks - for either bulk or quality recording. The pros and cons are reversed from those shown above. Therefore, if you select G.729A rather than G.711 the implications are: For All Recording A fraction of the bandwidth will be needed between the media processing resources and the recorder. More Avaya media processing resource will be needed. For Bulk Recording the capacity of a given server will be about doubled the quality of recordings will be a little higher the storage space used will be halved For Quality Recording the quality of recordings will be a little lower the storage space used will be one eighth (This is for voice recordings. Note that when combined with screen recordings, the overall impact is proportionately less). 42 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

43 Storage Requirements Storage Requirements Having determined the type, size and location of your recording capacity, you must now determine how much storage is required within the system. Storage requirements of many terabytes are not uncommon. Storage is required for: the operating system and applications installed on each server the audio content of recordings the database holding the searchable details of these recordings In each case you should consider requirements locally at each recorder and centrally. To do this, you will need to know: how many recordings will be made on the recording channels already identified (typically expressed in channel hours per day). In other words, how many hours of audio you expect to record on each recorder every day. how long you wish to retain recordings (typically expressed in days). the average duration of a call (typically expressed in seconds). Storage at Each Recorder In addition to the operating system, installed software and its configuration data, each recorder stores: details of the recordings it has made in a local database the recordings it has made as files on its hard disk (except for those controlled by equality Balance which are copied off to that server's storage as the call completes). In all cases, the system automatically notes the location of the recordings so that when a user wishes to replay a call that is no longer in local online (disk) storage, the application prompts them to insert the appropriate removable media (if any). Recordings are stored in G.726 compressed form at 16kbps (if received as G.711) or as G.729 files (8kbps) if received as such from the media processing resource on the switch. This allows for high volumes of recordings to be stored on the available disk space. Verint strongly recommends RAID arrays or fault tolerant Storage Attached Network (SAN) devices for online storage of recordings on the recorder platform. The recorder automatically manages the available recording storage space. This is used as a circular buffer providing instant access to the most recent recordings and deleting the oldest, as space is required for new recordings. Issue 5 May,

44 Planning and Prerequisites Rather than storing bulk recordings on RAID arrays in each recorder, many customers prefer to use Storage Attached Networks. These must be connected directly to recorders. Hierarchical File Storage (HFS) Systems and Network Attached Storage (NAS), however, can only be supported via the separate Archive application. The table below summarizes the requirements for Bulk Recording on a Verint ContactStore for Communication Manager. Details of Recordings Content of Recordings Stored In local postgres database Stored As WAV files on local disk in a hierarchical folder structure Purged Nightly, after user defined period (default 60 months) 100 at a time as disk space is needed for new recordings Volume Approximately 2KB per recording. Location Type So total ~ 2KB x recordings/day x days retained /var (in addition to the normal contents of /var) RAID 1 (mirrored) or 5 (striped) strongly recommended. Local hard drive or SAN, not NAS. Approximately 7.2MB (G.726) or 3.6MB (G.729) per channel hour of audio. So total ~ 7.2MB (or 3.6MB) x channel hrs/day x days retained /calls RAID 1 (mirrored) or 5 (striped) strongly recommended. Local hard drive or SAN, not NAS. Quality Monitoring When used to provide selective audio recording services for the Verint equality Balance application, the Verint ContactStore for Communication Manager only uses a small amount of local storage for voice recordings (less than 1GB). As each Quality Monitoring recording is completed, the recorder transfers it to the Quality Monitoring application's storage area. As the volume of recordings made by a selective, quality monitoring application is normally much less than that of a bulk recording system and is comparable to the screen recording volumes, audio is normally stored in G kbps PCM µ-law format (approximately 27MB per channel hour). This application also has to store: screen content recordings 44 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

45 Storage Requirements the agent configuration and evaluation data See Verint equality Balance (V5 or V7) Server Infrastructure Guide for further details of how to size the storage requirements for the Quality Monitoring application. Central Database Storage You should deploy Viewer or a Central Replay Server for search and replay if any of the following apply: you have more than one recorder and wish to search and replay from any recorder through a single query. you anticipate more than five concurrent users of the integral search and replay application on a Verint ContactStore for Communication Manager. If you wish to use Archive, you must install Viewer rather than a Central Replay Server. If you choose to deploy Viewer, you will need to size the storage it requires in accordance with Viewer Installation Guide to hold the call details from all the recorders that are contributing to it. For a Central Replay Server, follow the guidance above to size the storage needed for the call details using the total call volumes across all recorders that are feeding into the Central Replay Server. Archive Call Storage Where you wish to retain calls for longer than it takes to fill the hard disk storage on a recorder, two options are available: connect a DVD+RW drive to the recorder. All recordings made on the recorder will automatically be archived to the removable media on this (single) drive. This requires only 1GB of buffer space on the recorder. It is only suitable for systems recording less than 4GB per day. add Archive (an extra cost option) to the system. This application supports selective archival of recordings to multiple destinations. See Archive Insallation Guide for guidance on sizing of the storage for such systems. Backup Storage See Backup/Restore on page 141 for a discussion of Backup and Restore options. You should determine whether additional storage space is required in your corporate backup system to accommodate the new recording system. Issue 5 May,

46 Planning and Prerequisites Server Platform Taking the above factors and the potential location(s) of your recorders into consideration, you must determine how many channels of each type of recording you wish to deploy and on which site. Having decided the total recording capacity at each of your locations, you must translate this into one or more server platforms capable of handling the load identified. Dedicated server(s) must be provided with no other applications running on them. Sizing The table below shows the maximum capacity of four server types that are used as benchmarks for planning recorder hardware. Server Type Minimum A B C Server Specifications See also Platform Prerequisites on page 81 for detailed configuration pre-requisites. 2.0GHz single-processor Pentium IV with: 500MHz front side bus 512K of L2 cache 1GB RAM Hard disk with at least 7200rpm rating 100MBps full duplex Ethernet NIC port CD-ROM compatible drive for installation of software DVD+RW drive if local archiving is required. Sound card recommended. RedHat Enterprise Linux 4.0 or 5.0 Same as Minimum, but with: 3.0GHz single-processor Pentium IV Xeon with hyper-threading Dual 100Mbps full duplex Ethernet NIC ports RAID 1 or 5 strongly recommended Same as A, but with dual processor Same as A, but with quad processor 46 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

47 Server Platform Recorder The following table shows the maximum channel counts supported for the above servers. This table takes the following factors into account: Recording G.711 calls and compressing these to G.726 loads the server more heavily than receiving and storing G.729A calls. The CTI feed required for Conferenced mode increases the load on the server Using recorder ports for replay via the telephone or live monitor not only loads the recorder but these ports are more susceptible to packet jitter and therefore the recorder cannot be loaded as heavily when using these modes. These figures assume NO encryption is being used. If the audio to and from the Avaya system is encrypted, reduce capacities by 10%. If using a Key Management Server and encrypted storage, capacities are futher reduced. Exact figures are not yet available. Maximum Channel Count G.711 from Communication Manager G.729A from Communication Manager No Telephone Replay or Live Monitor With Telephone Replay or Live Monitor No Telephone Replay or Live Monitor With Telephone Replay or Live Monitor Server Conf Others Conf Others Conf Others Conf Others Min. N/R 60 N/R N/R N/R 120 N/R N/R A (Single Xeon) B (Dual Xeon) Conf = Conferenced mode recording ports. Others = All other modes. N/R = Not recommended. Sizings assume an 8 hour working day. For 24 x 7 running, provide one server level higher (for example, use server C instead of B). Sizings assume moderate replay load. See below for further details. Sizings assume an average of a three minute call handle time. Issue 5 May,

48 Planning and Prerequisites Maximum Channel Count G.711 from Communication Manager G.729A from Communication Manager No Telephone Replay or Live Monitor With Telephone Replay or Live Monitor No Telephone Replay or Live Monitor With Telephone Replay or Live Monitor Server Conf Others Conf Others Conf Others Conf Others C (Quad Xeon) Replay Load When using Replay Ports there are additional considerations. The above sizing assumes a moderate replay load of up to 5 concurrent replays. If using telephone replay, up to 8 ports can be supported (of which rarely will more than 5 actually be playing concurrently) without compromising these loading limits. To support up to 10 concurrent replays, increase the server specification one level (for example, use server C instead of B). For higher loads, separate replay onto a dedicated Viewer or Central Replay server. Central Replay Server Use server type: NOTES Conf = Conferenced mode recording ports. Others = All other modes. N/R = Not recommended. Sizings assume an 8 hour working day. For 24 x 7 running, provide one server level higher (for example, use server C instead of B). Sizings assume moderate replay load. See below for further details. Sizings assume an average of a three minute call handle time. B for a Central Replay Server supporting up to 1000 recording channels on the recorders feeding it and up to 40 telephone replay ports. Consult Verint in-region sales for larger scale topologies. 48 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

49 Server Platform DVD+RW Drive A wide range of DVD drives has been used successfully and Verint is not aware of any specific model limitations at this time. Only single-layer 4.7GB DVD+RW media are supported. Note: Note: Most drives are advertised as supporting a wide range of subtly different media types (e.g. DVD+R, DVD-R, DVD+RW, DVD-RW, single and double layer etc.). Regardless of which media the drive supports, the recorder ONLY supports writing of DVD+RW single layer media. You must confirm that the drive supports this media type, works under the version of operating system you are using and that you only insert this type of media. RedHat 5 RedHat Enterprise Linux 5 incorporates new features as part of it s Hardware Abstraction Layer. You need additional steps to turn off the HAL s media detection service, which interferes with the archiver. To disable the media detection service follow these steps: 1. Execute the "lshal" command and direct its output to a file lshal > hal.txt 2. Open the created file and look for the string storage.drive_type. It will list all devices, local and external, so you will find the floppy, cdrom etc. 3. Locate your device. It would be either cdrom, cdrom1..etc, or dvd. The adjacent lines for storage model and vendor will contain specific details regarding the device, for example: storage.serial = 'HL-DT-ST_DVDRAM_GSA-E50L_P ' (string) storage.vendor = 'HL-DT-ST' (string) storage.model = 'DVDRAM GSA-E50L' (string) storage.drive_type = 'cdrom' (string) Issue 5 May,

50 Planning and Prerequisites 4. Scroll up until you find the line starting udi = The udi is likely to be in the form /org/freedesktop/hal/devices/storage_serial_vendor_model_serial number For example: Udi='/org/freedesktop/Hal/devices/storage_serial_HL_DT_ST_DVDRA M_GSA_E50L_P ' 5. Test the value of the media_check_enabled flag using the hal-get-property command. Substitute your actual UDI into the commands: hal-get-property --udi /org/freedesktop/hal/... --key storage.media_check_enabled the result is likely to be "true" - it needs to be "false" 6. Set it to false using hal-set-property command: hal-set-property --udi /org/freedesktop/hal/... --key storage.media_check_enabled --bool false 7. Re-check that the value is now "false" using the command in step Add the command from step 6 into the file /etc/rc.local (This will run the command automatically for you every time the machine reboots.) 9. To find the soft link created for the device return to the text file, and look for the value for "block.device" in the Device's settings. It will look something like this: block.device = '/dev/scd0' (string) 10. Enter the location in the recorder under Local DVD+RW Drive (on the Server page) 50 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

51 Network Issues Network Issues In planning the network that will support your Verint ContactStore for Communication Manager system, you must consider: the additional load imposed on the network the IP ports used - so that firewalls can be configured appropriately Load You must design your network topology to accommodate the additional traffic created by the recording system. Each active recording port places the same load on the network as an IP phone carrying the same call. Note that the recorder uses a 60ms packet interval, which significantly reduces the total network load. Ports Used The components of the system use a number of IP ports to communicate: between each other with various Avaya Communication Manager components with end-users and administrators Recorder Interfaces on page 199 provides a diagram and table listing all of the interfaces to and from the Verint ContactStore for Communication Manager software. Your network and firewalls must be configured to permit traffic to pass over these links. Network Address Translation Routing The IP address of a Verint ContactStore for Communication Manager is sent to the AE server as part of the softphone registration. It is therefore essential that when the media processing resources transmit to this address, the packets are routed correctly to the recorder. The recorder must therefore be visible to the media processing resources on the IP address that it uses itself. Additionally, if the recorder has two NIC cards it is imperative to ensure that all VoIP packets are transmitted over the same NIC card (i.e. the network route to all Media Processors must be the same). Issue 5 May,

52 Planning and Prerequisites Recorder Licensing Verint ContactStore for Communication Managers are licensed through an Activation Key which is tied to the hardware for which is it is issued. The license key contains details of server type and port counts. It may also contain an expiry date. Server Licenses You will need: a server license for each Verint ContactStore for Communication Manager a Central Replay Server license for each Central Replay Server Port Licenses You will need an appropriate "port license" for each port on a recorder. There are five types of port licenses as discussed under Recording Functionality on page 25. The table below summarises which licenses are required for each recording mode. Port Allocated to Station Executive Recording On Demand Recording Meeting Recording Station Bulk Recording Conferenced Recording Externally Controlled Recording equality Balance V5 (Recording or Replay) Telephone Replay Live Monitor Port License Required Part-time Recording (though a full-time recording license can be used if available) Full-time Recording Quality Replay Live Monitor 52 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

53 Recorder Licensing Other Licenses You will also require Avaya licenses as described in DMCC (IP_API_A) Licenses on page 60 and TSAPI Simultaneous User Licenses on page 60. Timed Trials Verint can, at its discretion, issue an activation key which includes an expiration date. This allows for timed trials of any combination of features and capacity. As the expiration date is fixed within the license, the server will stop operating at that date regardless of when you enter the license key. To extend a timed trial of to upgrade to a full license, contact Verint for an updated license activation key. A five day trial license is available automatically from the license key entry page.! CAUTION: CAUTION: The five day trial license must not be used for production recording. When a full license is installed, any trial recordings become unplayable Issue 5 May,

54 Planning and Prerequisites Avaya system prerequisites To use the Verint ContactStore for Communication Manager system you will need to ensure that your Avaya system meets the following requirements. This section discusses the various hardware, software and configuration requirements. Communication Manager Model Station Count Loading Software Version Verint ContactStore for Communication Manager requires AE Services and hence is only supported on the models and versions of Communication Manager that support this platform. The Avaya media server running Communication Manager must be an S8xxx system (currently S8300 through S8720). Each port on a recorder is an additional IP phone on the switch. Do not exceed the total station count for the switch in question. Each recording adds as much load to a switch as a normal call. Hence you can only record 100% of calls if your switch is running at no more than 50% of its design load. For example, S8700 switches running at up to 20,000 BHCA (complex call center call types) can have all calls recorded. Higher loads would require an S8710 or S8720. Verint ContactStore for Communication Manager requires one of the following: CM 3.1- CM 4.0 Please check on for more recent loads. 54 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

55 Avaya system prerequisites Gateway Resources Card Slots Time-slots These house the media switching components of the Avaya system. You must ensure that the system has, or is expanded to have sufficient: Card Slots for the C-LANs and Media Processing Resources described below Time-slots for the original calls and, where needed, the recording channels. Each C-LAN and Media Processing card must be located in the appropriate gateway and therefore in an available card slot alongside the existing cards. The recorder conferences into calls in order to record them - using Single-step conference (conferenced modes only) or service observe. ANY recording in which the recorder is injecting beep-tone will require one additional time-slot per concurrent recording. With Service Observe modes, use the "No-talk" option (if your switch supports it) in preference to the "Listen-only" variant. The latter requires an additional time-slot per concurrent recording - even if you do not inject beep tone. Where additional timeslots are needed, the total timeslot count must not exceed the maximum available on that port network (484). Therefore, for a 100% recorded system, using service observe on a switch that does not support "No-talk" observe, or beep tone injection do not equip any port network with more than: 6 x T1s (=144 calls, 432 timeslots) or 5 x E1s (=150 calls, 450 timeslots) These guidelines allow for reasonable additional timeslot usage for conferences with other port networks, shared tones and so on. Rebalance port networks or add new ones to reduce the timeslot requirement on each to this level. AE Services Each port on a Verint ContactStore for Communication Manager uses a DMCC softphone. When using Conferenced recording mode (or when an external controller needs to instruct Issue 5 May,

56 Planning and Prerequisites Loading the record to single-step conference into calls) the recorder also makes use of Avaya CT services. From Communication Manager 3.0 onwards, these are both provided by Avaya's AE Services platform. Note: Note: To avoid overloading an AE Server, do not attempt to record more than 20,000 calls per hour through each AE Server. (20,000 BHCA). Note: Note: You must not use more than 1,000 softphones (recorder ports) through a single AE Server. If several small recorders are used, you may connect them to a shared AE Server, but only if the total load on the AE Server does not exceed this figure. If the load imposed by a single recorder exceeds this figure, you must split the load across multiple smaller recorders, spreading the load across multiple AE Servers. Multiple AE Servers Location Vintage Most Communication Managers can support up to 15 AE Servers but this is a total count - not just those associated with recording. You may have othe AE Servers associated with other CTI applications. In a multi-site system, you should always aim to install an AE server on the same site as the recorder(s) that is (are) using it. This minimizes the chance of system failure due to loss of connectivity between recorder and switching system. Verint ContactStore for Communication Manager 7.8 has been tested against AE Services 4.0. Ensure you are running the latest recommended load of AE Services. Expansion Interface Boards (TN570) All Expansion Interface Boards must be TN570C Vintage 3 or later. 56 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

57 Avaya system prerequisites C-LAN C-LANs (TN799 DP) are used for two purposes: Number of C-LANs CTI information is passed through them DMCC softphones register through them. To ensure that a C-LAN does not become a single point of failure in a recording system, you should always provide at least two C-LANs for each AE server. As the CTI load and channel count increases, you should provide more C-LANs as shown below. C-LANs per AE Server Maximum BHCA through the AE Server Maximum Recording Channels through the AE Server 1 NOT SUPPORTED NOT SUPPORTED 2 12, , ,000 (AE Server limited) ,000 (AE Server limited) 800 Location of C-LANs Vintage Firmware For maximum resilience, spread C-LANs across multiple port networks. To avoid bottlenecks between the port network and the switch, do not connect more than 400 recording ports to the C-LANs in any port network. Refer to the switch documentation for the release of Communication Manager you are running. The latest update is recommended but these cards must be at least at Firmware update 132. Issue 5 May,

58 Planning and Prerequisites VoIP Resources Each port on the recorder will use media processing resources on the Avaya system when it is active (recording or replaying via the telephone). You must ensure that sufficient media processing resources are available for the recording and replay load - in addition to any existing use of these resources within your system. Resource Requirements G.711 recording or replay uses less resource than G.729A recording. Note that replay is always performed using G.711. Depending on the type and version of your Communication Manager, you may require one or more of the devices shown in the table below. Resource Type Comments G.711 Recording or Replay Ports G.729A Recording Ports Media Processing Resource TN2302AP MM760 VoIP Module Media Processing Resource 2602AP Included within S These requirements are solely for the recorder s ports and are in addition to any used by other IP phones or other switch components. Note: Note: It is not recommended to dedicate Media Processing resource to recording so it is important to over- rather than under-provide as other users of this resource could otherwise starve the recorder of this capability. On the plus side, you may use existing spare capacity in the switch for recording - but check the location of the resource as well as the amount. Location of Resources When adding recording to systems with multiple port networks, it is vital to check that the recordings do not overload the interconnects between port networks. If a call cannot be recorded using VoIP resources within a port network that the call would have been routed through anyway, then the call must be routed to another port network to 58 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

59 Avaya system prerequisites reach the VoIP resource. This varies according to whether the phones are digital (DCP) or IP and, with IP phones, whether the system is IP- or Multi-connect based. Site the VoIP resources according to the table below. Recording System DCP Phones IP Phones Multi-Connect IP Phones IPConnect Station-side - High % of calls on trunks recorded. (Station Bulk, Station Executive or Unify applying station-based rules.) Same Port Network as the Phones being recorded. N x VoIP resources per phone being recorded. Same Port Network as the trunks carrying the calls. N x VoIP resources per trunk channel on that port network that could be recorded concurrently. Station-side - Low % (<25%) of calls on trunks recorded.(station Bulk, Station Executive or Unify applying station-based rules.) Dedicated port network(s). VoIP resources equal to N x total number of trunk channels that could be recorded concurrently As above Where, N=1 for G.711 recording and N=2 for G.729A recording. Vintage Firmware Fault Tolerance Refer to the switch documentation for the release of Communication Manager you are running. The latest update is recommended but Media Processors must be at least at Firmware update 105. You should consider providing one additional board per port network. In the event of a board failing, a spare is then available to handle the full recording load, without having to look outside the Media Gateway - which could introduce sub-optimal use of back-plane timeslots and potentially impact recording in other gateways. Issue 5 May,

60 Planning and Prerequisites Further Information For more information, refer to Chapter 2: Administering C-LAN and IP Media Processor circuit packs, in the Administering Converged Networks section of the Administration for Network Connectivity for Avaya Communication Manager manual. Multi-Connect Capacity Keeping in mind the number and location of recorders and VoIP resources as defined above, confirm that the capacity of the Multi-connect switch (if present) is not exceeded. DMCC (IP_API_A) Licenses Each port on a recorder requires one Avaya IP_API_A license. TSAPI Simultaneous User Licenses All recording modes require that Avaya CT be enabled, but only one license is needed. (DMCC s Call Information Services are provided by Avaya CT.) Conferenced recording modes (only) require Type A licenses as follows: Purpose For AES to connect To record calls To track agent logins Licenses Required 1 Maximum Number of concurrent recordings Number of Skill groups being observed to track logins To observe stations currently of interest If recording specific Stations Number of stations to be recorded If recording specific agents Maximum number of these agents logged in concurrently If recording specific skill groups Maximum number of these agents logged in concurrently If recording specific VDNs 0 60 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

61 Avaya system prerequisites Note: Note: If an external controller is tracking CTI events, then it may require Avaya CT licenses. In addition to its requirements, the recorder will still need one license per concurrent recording and AES will still need its 1 license. VoIP Network Design The recorder hosts the equivalent of 1 x Avaya 4624 IP Phone per port - whether recording, replay or live monitor. You must therefore design the connectivity between it and the rest of the Avaya switch infrastructure as if there were a bank of this many IP phones at the location of the recorder. Follow Avaya's network design guidelines for this number of IP softphones operating in either G.711 or G.729A mode, but with 60ms packet intervals. If the bandwidth between recorder and the media processing resources it uses is less than LAN speeds (100Mbps full duplex) then you should use G.729A recording only. Issue 5 May,

62 Planning and Prerequisites Topologies This document has so far discussed functionality in terms of "applications" without being specific as to the physical location of these. As the individual components of the recording system interconnect using IP-based mechanisms, you may distribute the components across your Enterprise's network in a wide range of topologies. In small systems, a single server can perform recording, storage and retrieval but in larger systems, you can separate these tasks onto different servers in a variety of ways as discussed below. A number of the basic topologies have already been shown in the diagrams of Recording Methods on page 31. The following paragraphs define the rules under which each of these basic topologies can be used and introduce the more advanced topologies required for larger and more fault tolerant systems. Bulk Recording System At its simplest level, a Verint ContactStore for Communication Manager system consists of a single linux server running the Verint ContactStore for Communication Manager software and configured as a Master. This application provides: system administration functions via a browser voice recording of Avaya phonesets using DMCC IP softphones integration to the Avaya Communication Manager via Application Enablement Services (AES) for real-time control and tagging of recordings integral search and replay capabilities. See the diagram in Introduction on page 16 for this basic Bulk Recording Topology For most small to medium sized bulk recording requirements, this single server is all that is required. Sampled Quality Monitoring (only) System When used for sampled Quality Monitoring, an additional software application, Verint equality Balance, is required. While the underlying Verint Contact Recording Master provides the voice recording capabilities of the overall system, this optional application adds: evaluation form design various call selection methods 62 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

63 Topologies screen content recording and storage agent performance recording and reporting Bulk Recording + Quality Monitoring System Bulk recording and Quality recording can be combined in two different ways: 1. Selective Quality Recording + Bulk - in which the quality application chooses to record specific calls. These recordings are made in addition to the bulk recordings and the resultant files are copied across to the Quality Monitoring application's file share as each recording ends. If the bulk recordings are being made using Service Observe, the Quality recordings are "split" off from these within the recorder - so as NOT to load the Avaya system any more. If, however, the bulk recordings are mad using conference methods, the selective quality recordings (which always use service observe) will also load the Avaya switch. 2. "ContactStore Plus" mode - this is where the bulk recording system advises eqaulity Balance (R7 only) of all the calls that it has recorded. The Quality application does not ask for additional recordings to be made. It will simply identify a subset of the bulk recordings that it will make available for evaluation. In this case the recordings used for evaluation are the same recordings made for bulk recording. Large Bulk Recording Systems Where your recording load exceeds the capacity of a single server, or where a distributed topology is more appropriate, you may deploy multiple recorders - in one of two ways. Partitioned Systems If your recording requirements can be completely separated, you can deploy multiple independent recorders, each unaware of the others. This applies if you are using: Station Bulk mode Station Executive mode Conferenced mode to target specified stations. Master/Slave Recorders Where you are using Conferenced mode to record on the basis of Agent ID, Skill hunt group or VDN, there must be a single recorder in charge of all recordings. To support large systems, you can split the control of recording from the actual recording of the audio stream. Issue 5 May,

64 Planning and Prerequisites In such cases, the "Master" recorder is connected to the Avaya CT feed and is aware of the recording rules - and of the type and locations of the other, "Slave" recorders as shown below. The Master recorder communicates with each Slave via a TCP/IP link. It instructs the slaves to tag the required data/voice streams with the details it learns from the Avaya CT link. Recorder Type and Location You may distribute recorders across your network. This lets you trade off network load versus security of storage. For example, if you wish to record calls on an overseas site to which you have limited bandwidth, you can locate a recorder and media processing resources on that site. High Availability Systems Because the recording system is based on industry standard PC hardware, you can spend as much or as little as you like on fault tolerant hardware to increase the reliability of each server. Verint recommends that you use fault tolerant, hot-swappable RAID disk arrays, redundant power supplies and fans as a matter of course. 64 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

65 Topologies Bonded and/or dual NIC cards are strongly recommended. Where two independent NIC cards are used, these should be connected to independent Ethernet switches to provide maximum fault tolerance. (Never give the two interfaces IP addresses in the same subnet.) For still higher availability, you can provide one or more standby recorders. Advanced configuration options let the recorders match your switch failover modes - supporting both ESS and LSP. This is a complex topic and a brief summary of the options is given below. (See also Fault Tolerant Systems on page 251 for further details of how to design and configure fault tolerant systems) Master/Standby Recorders In small systems, you can install and configure a Standby recorder to "shadow" another recorder - referred to as the Master. The standby copies configuration details automatically from the primary via a fault-tolerant TCP/IP connection between the two. Over these same links, the two recorders exchange "heartbeats" every second. The standby will take over should the heartbeat fail or should the master request that it do so. This latter case will occur, for example, due to the master s disk filling or connection to the AE Services failing. The figure below shows a typical configuration. Issue 5 May,

66 Planning and Prerequisites Note: Note: It is normal to provide Viewer or a Central Replay Server so that all recordings, whether made by Master or Standby, are visible to end users in a central database. Otherwise, users would have to know which recorder was active at a given time and search against its database to be sure of finding a recording. In larger systems, where a Master recorder controls one or more "Slave" recorders, the use of a Standby recorder ensures that the Master does not become a single point of failure for the whole system. In this case, each of the slaves permanently connects to both Master and Standby. This allows the Standby to take over rapidly should the Master fail. Fault Tolerant Pools When using Conferenced mode with a Master and one or more Slave recorders, recordings can be directed towards any available recorder. Therefore a "pool" of Slave recorders can be equipped with one more recorder than is needed for the projected loading. In this configuration, if one Slave fails, the Master recorder will reassign the calls being recorded by the failed unit across the remaining slaves. Centralized Applications In any system with more than one recorder, you may wish to deploy central search and replay (Viewer or Central Replay Server) and/or Archive applications. In this case, each of the recorders: provides details of its recordings to the central database responds to requests from the central applications for replay of recordings that it is holding. See Viewer Installation, Archive Installation and Archive Administrative Guides for sizing and planning of these servers. Note that the recording system is designed to continue recording regardless of the state of these central servers. This means that the availability of the central applications cannot affect the reliability of the recorders themselves. 66 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

67 Integrating with other systems Integrating with other systems The Verint ContactStore for Communication Manager system supports integration with a wide variety of other applications including other CTI feeds, third party and customers' own applications. There are three options, depending on the complexity of the integration required. Supplementary Tagging of Bulk Recordings It is a common requirement to "tag" recordings with additional details, such as customer account numbers, trouble-ticket numbers etc. This information is often held in a third party or custom application and is known while the call is in progress. The recorder implements a TCP/IP socket based interface. Applications can: receive events as recordings start and end send "tag" information to the recorder to be associated with the recording, alongside the basic call details. Users can then search for recordings based on this enhanced set of call details. This Recorder Control Protocol is very simple as the recorder makes contact with the application at an IP address specified in the recorder's administration pages. The recorder advises the application as recordings start and stop on each station. Each start message contains a unique reference (or "INUM") to that recording. If the application wants to tag a call, it sends a "TAG" command specifying the INUM and station number plus the data formatted as XML. In this mode, the recorder continues to be responsible for recording rules. Should the tagging application have a problem, the worst it can do is tag calls incorrectly. This interface is included within the core Verint ContactStore for Communication Manager software and you may use it immediately. A typical tagging project requires the purchase of 2-3 person-days of services to learn how to interface to and control the recorder. See External Control Interface on page 233 for further details of the Recorder Control Protocol, associated APIs and example applications that are available. Explicit External Control of Bulk Recording In normal operation, the recorder is given some basic Station/Agent/Skill Group/VDN rules, that it uses to determine which calls are to be recorded. However, you can build more sophisticated recording systems in which the decision to record is made by an existing call Issue 5 May,

68 Planning and Prerequisites routing/handling application. By including control of recording as an integral part of call flow, such applications can, for example, automatically stop recording a call during security questions and resume once the customer has been validated automatically start recording as the agent accesses the "customer complaint" form add additional tagging as the agent traverses a series of menus Applications can control recording using the same interface as described under Supplementary Tagging of Bulk Recordings on page 67. By using the additional commands "START", "STOP" and "BREAK", they can override the rules applied by the administration interface. This interface is included within the core Verint ContactStore for Communication Manager software and you may use it immediately. A typical external control project requires the purchase of five person-days of services to learn how to interface to and control the recorder. Enhanced Quality Monitoring The above sections primarily relate to Bulk Recording operation. To extend the scope of the recording rules within the Quality Monitoring application, additional inputs are connected in two ways. Using the "eqconnect" toolkit, you can provide additional data fields from other applications to complement those available from Avaya CT. Where a layered product (such as Genesys T-server) manages all calls, you can connect its CTI feed as an ALTERNATIVE to the CTI information received from Avaya CT. In this case, the Quality Monitoring application connects via Verint Systems' other CTI integrations to the CTI feed (in this case the Genesys T-server). The decisions about what to record for quality purposes and how to tag the calls are made by the Quality Monitoring application on the basis of this CTI feed - without input from the Verint ContactStore for Communication Manager Master. The Recorder simply acts as a "dumb" recording mechanism, recording the stations that the Quality Monitoring system requests in real-time. Both of these approaches incur additional licensing charges as well as the costs for the services required to install and configure these options. 68 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

69 Chapter 3: Installation This chapter gives details of the steps to install a Verint ContactStore for Communication Manager system. The main sections in this chapter are: Overview on page 70 Avaya System Configuration on page 71 Order in which to Install Applications on page 80 Platform Prerequisites on page 81 Installing Verint ContactStore for Communication Manager on page 86 Installing Verint equality Balance (V5) on page 88 Installing Verint equality Balance (V7) on page 89 Installing Viewer on page 90 Installing Archive on page 94 Issue 5 May,

70 Installation Overview Installation of a complete recording system requires: Configuration of your Avaya system to support recording. Planning the order in which to install the application servers Preparing each server Installing the Verint software on each server Installing client software for users of the Quality Monitoring application (if required). 70 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

71 Avaya System Configuration Avaya System Configuration Before installing any Verint ContactStore for Communication Manager components, you must ensure that your Avaya Communication Manager is correctly configured and, where necessary, upgraded to support the recording system. As you complete these steps you will be asked to note a number of details which you will need later when configuring the recorder. Prerequisites Refer to the appropriate Avaya documentation to apply any prerequisite upgrades and/or additional licenses as detailed in Avaya system prerequisites on page 54. Communication Manager Configuration Customer-options Use the Avaya administration interface to configure the following items: Set the required system parameters and check that IP_API_A licenses have been installed as follows: 1. Run the following command display system-parameters customer-options 2. On page 4, verify that Enhanced Conferencing is set to y. 3. On page 6, verify that Service Observing Basic and Service Observing Remote/By FAC are set to y. Note: Note: Station Bulk, Station Executive, Quality and (optionally) Unify Recording modes use Service Observe. 4. On page 9 or 10 (it varies depending on your system), check the Maximum IP registrations by product ID field. This field tells you how many IP_API_A licenses you have and how many you have already used. If you do not have enough IP_API_A licenses or if IP_API_A product ID is not listed on this screen, contact your Avaya representative for more licenses. Issue 5 May,

72 Installation Features You must set the following system-wide CM parameters. 1. Enter the following command line: change system-parameters features 2. On Page 5, set Create Universal Call ID (UCID) to y and allocate a number to the switch if it does not already have a unique reference. If there is only one switch, set it to On Page 12, set Send UCID to ASAI. Adding IP softphones You must add a station on the Communication Manager for each port on the recorder. Create all stations identically. You will subsequently use the recorder's Administration pages to assign them to the various modes. 1. Use the add station command to add as many stations as there are ports on your recorders. Note the station numbers as you will need to enter these into the recorder later. 2. Run add station xxxx, where xxxx is the new station's extension that you want to administer. 3. For Station Type, enter For Security Code, enter the numeric security code the recorder must use to register softphones with the Communication Manager. Note this security code as you will need to enter it into the recorder later. 5. Verify that the Class of Restriction (CoR) you choose has Can Be A Service Observer enabled and Can Be Service Observed disabled when you intend to use any of the following recording modes: a. Station Bulk b. Station Executive c. Quality d. Unify/External Control (Service Observe) Important:! Important: Verint recommends that you use a dedicated CoR for the stations being used as ports on the recorder. The CoR of the stations you want to record with the recording modes mentioned above must have Can Be Service Observed enabled. For more information on how to administer Class of Restriction and Service Observing, refer to the following resources: 72 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

73 Avaya System Configuration WARNING: Class of Restriction Section, in Chapter 19: Screen Reference of the Administrator Guide for Avaya Communication Manager. Service Observing topics, Feature Access Code and Feature Related System Parameters Sections, in Chapter 19: Screen Reference of the Administrator Guide for Avaya Communication Manager. Service Observing, in the chapter titled ACD contact center features of the Avaya Communication Manager, Release 3.0, Call Center Software Automatic Call Distribution (ACD) Guide.! WARNING: Since a station can only be service observed once (CM 3.x) or twice (CM 4.x), it is important that you disable, through the use of appropriate CoRs, the ability of anyone else to service observe those stations that must be recorded at all times. If you must listen to these stations while they are being recorded, use additional Live Monitor ports on the recorder. 6. In the IP Softphone field, enter y. 7. Set Display Language to english. 8. On Page 2: Set IP-IP Audio Connections and IP Audio Hairpinning to n 9. On Page 3: a. Assign the following feature buttons in addition to the three default call appearances: i. Button 4: conf-dsp ii. Button 5: serv-obsrv b. Clear the call-appr setting on Button 3. Issue 5 May,

74 Installation change station Page 1 of 4 STATION Extension: Lock Messages? n BCC: 0 Type: 4624 Security Code: TN: 1 Port: S00081 Coverage Path 1: COR: 1 Name: CCE Line 01 Coverage Path 2: COS: 1 Hunt-to Station: STATION OPTIONS Loss Group: 19 Personalized Ringing Pattern: 1 Message Lamp Ext: Speakerphone: 2-way Mute Button Enabled? y Display Language: english Survivable GK Node Name: Survivable COR: internal Media Complex Ext: Survivable Trunk Dest? y IP SoftPhone? y IP Video Softphone? n change station Page 3 of 4 STATION SITE DATA Room: Headset? n Jack: Speaker? n Cable: Mounting: d Floor: Cord Length: 0 Building: Set Color: ABBREVIATED DIALING List1: List2: List3: BUTTON ASSIGNMENTS 1: call-appr 7: 2: call-appr 8: 3: 9: 4: conf-dsp 10: 5: serv-obsrv 11: 6: 12: Administering hunt groups If you want to use On Demand Recording or Meeting Recording modes, consider grouping these ports into one or more hunt groups for each mode. Users can then access the recording functionality through the number of the hunt group rather than through individual ports. 74 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

75 Avaya System Configuration You could assign all ports for a recording mode to a single hunt group to provide a single shared pool of ports, available on a "first come, first served" strategy. Alternatively, you could split the pools into several independent hunt groups - or even leave some individual ports. For example, a dedicated port to be used only by the conference phone in the board room would ensure that meetings there could always be recorded. For more information about hunt groups, refer to Managing Hunt Groups in Chapter 7: Handling incoming calls, in Volume 1 of the Administrator Guide for Avaya Communication Manager. Note which ports you have assigned to which hunt groups. Configuring tone detection The recorder uses tone detection for Meeting Recording, Station Executive Recording and Live Monitoring. If you specify a delete command for Station Bulk recording, it is also used in this mode. To configure tone detection: 1. Type change system-parameters ip-options. 2. In the Intra-System IP DTMF Transmission Mode field on Page 2, enter rtp-payload change system-parameters ip-options Page 2 of 2 IP-OPTIONS SYSTEM PARAMETERS Always use G.711 (30ms, no SS) for intra-switch Music-On-Hold? n IP DTMF TRANSMISSION MODE Intra-System IP DTMF Transmission Mode: rtp-payload Inter-System IP DTMF: See Signaling Group Forms Network Region setup The recorder requires the following: The DMCC softphones used as recorder ports must be in an IP network region that supports G.729A and G.711MU with 60ms packet intervals and NO OTHER CODECS. There must be a media gateway or a media processor resource in the same network region or in an interconnected network region. To set up a network region as above: Issue 5 May,

76 Installation Create a Codec Set Create a new codec set specifically for the recorder(s) as follows: 1. Choose an unused codec set number for the recorders 2. Use the change ip-codec-set setnumber command to create a codec set that uses G.729A and G.711MU 3. Verify that G.711MU and G.729A are the ONLY codecs in the codec set. 4. Set Silence Suppression to n. 5. Set Frames Per Pkt to 6 - which will show a Packet Size of 60 (ms). 6. Set the first two Media Encryption options to none and aes respectively. 7. Ensure that FAX, Modem, TDD/TTY are all set to off. 8. Ensure that Allow Direct-IP Multimedia is set to n. change ip-codec-set 4 Page 1 of 2 Codec Set: 4 IP Codec Set Audio Silence Frames Packet Codec Suppression Per Pkt Size(ms) 1: G.711MU n : G.729A n : 4: 5: 6: 7: Refer to the following for further information: For an explanation of administering IP codec sets, refer to the Administering IP Codec Sets Section, in Chapter 4: Network Quality Administration, of the Administration for Network Connectivity for Avaya Communication Manager guide. For a screen reference, refer to the IP Codec Set Section, in Chapter 19: Screen Reference, of the Administrator Guide for Avaya Communication Manager. Create a network region Create a new network region and assign the previously created codec set to it as follows: 1. Choose an unused network region for the recorders softphones 2. Type change ip-network-region region where region is the number of the chosen network region. 76 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

77 Avaya System Configuration 3. Specify the Codec Set created in the previous step. 4. Set the two IP-IP Direct Audio options to No. 5. Set IP Audio Hairpinning to n. 6. Complete the region-region codec table with the number of the Codec Set created in the previous set. Refer to the following for detailed information: Administering IP Network Regions Section, in Chapter 4: Network Quality Administration of the Administration for Network Connectivity for Avaya Communication Manager guide. For a screen reference, refer to the IP Network Region Section, in Chapter 19: Screen Reference of the Administrator Guide for Avaya Communication Manager. Assign softphones to the network region To ensure that the recorder's softphones register in the network region created above, use the change ip-network-map command. Add the IP address of the AE Server to the new network region so that all the recorder's softphones - which register via that AE Server - are created in this network region. The following example shows how to complete the form. In this example, the AE Server is using network region 1 with an IP address of change ip-network-map Page 1 of 32 IP ADDRESS MAPPING Emergency Subnet Location From IP Address (To IP Address or Mask) Region VLAN Extension n n n n Recording of Bridged Line Appearances Where a bridged line appearance appears on multiple physical sets, and calls made on the bridged line need to be recorded, you must enable Service Observe Physical Set. This is also know as the SA7900 Special Feature. Configure Station Bulk recording to target the main line appearance on each of the physical sets. When a call is answered on the bridge line, the special feature allows service observe to observe the bridged appearance call. Issue 5 May,

78 Installation AE Server Configuration The AE Server provides the recorder with DMCC client services and, (if using Conferenced recording or equality Balance in selective mode) TSAPI services. The following instructions assume the AE Server has been installed specifically for recording and that OAM and User Management administrative accounts have been created. If the AE Server to be used is already in use for other purposes, check these settings and confirm that its current configuration is appropriate. Administer C-LANs Use the AES Administration Screens to add the C-LANs that will be used to register softphones. 1. From the CTI OAM Admin OAM main menu, select: Administration > Switch Connections 2. Click Edit H.323 Gatekeeper. OAM displays the Edit H.323 Gatekeeper - Switch1 page. 3. In the Name or IP Address field, type the hostname or IP Address of the switch C-LAN, and then click Add Name or IP. TSAPI Configuration User Account Even if there are no explicit CTI clients, you must configure TSAPI (also known as Avaya CT). Do this after AE Services have been configured as above. If you are using the Security Database for Authentication, create a CTI user account on the AE Server as follows: 1. Go to: User Management > Add User 2. Complete all of the required fields (indicated by an asterisk). 3. Select userservice,useradmin from the Avaya Role drop-down menu. 4. Select Yes from the CT User drop-down menu. 5. Ensure that the new CTI user has access to all TSAPI-controlled devices by going to: Administration > Security Database > CTI Users > List All Users 6. Click Enable next to the Unrestricted Access option. 78 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

79 Avaya System Configuration Test Phonesets You should provide three Avaya phones close to the recorder - in the same or a neighboring rack. Configure these phones with all of the features in use on the phones that you intend to record. You can then use these to place test calls while working on the recorder. Issue 5 May,

80 Installation Order in which to Install Applications In many cases, the recording system will consist solely of a single Verint ContactStore for Communication Manager Master server. However, in topologies that are more complex it is important that you install the basic recording infrastructure first and then layer the other applications on top. Note: Note: If you have multiple independent recorders, install one recorder first, then the central applications. Finally install the other recorders. This way you will address any problems with the recorder installation before you have repeated them many times. Install server components in the order shown below: 1. Install one Verint ContactStore for Communication Manager (the Master if using Master/Standby or Master/Slave). Follow the procedures in Platform Prerequisites on page 81 and Installing Verint ContactStore for Communication Manager on page Install Verint equality Balance (V5 or V7) (if required) in accordance with the appropriate Installation Guide and configure the Verint ContactStore for Communication Manager to support it as described in IInstalling Verint equality Balance (V5) on page 88 or Installing Verint equality Balance (V7) on page Install Viewer as described in Installing Viewer on page Install Verint Contact Archive (if required) in accordance with Archive Installation Guide and configure it to support the recorders as described in Installing Archive on page Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

81 Platform Prerequisites Platform Prerequisites Before installing the Verint ContactStore for Communication Manager (Master, Slave or Standby) software on the designated server(s), you must prepare each server as described below. Note: Note: For the other server components refer to the relevant product manual rather than this one. Operating System Version Disk Storage The operating system must be RedHat Enterprise Linux Version 4.0 Update 5 or higher or Version 5.0. RedHat Enterprise Linux Version 3.0 is not supported for new installations but existing systems running on it can be upgraded to CSCM 7.8. The operating system must be installed using the RedHat kickstart process. Verint supplies a tool to generate the kickstart script automatically. If you cannot use the kickstart process you must contact Verint Professional Services for guidance. You must plan the partitioning of your server's disk(s) in line with the storage needs outlined in Storage Requirements on page 43. The kickstart script supports servers with one logical disk or two physical disks. If using RAID, use the RAID utility to make one large logical volume. The kickstart script partitions the disk(s) as follows: Mount Point Use One Physical or Logical Disk Two Physical Disks /boot Bootstrap 100 MB 100 MB on first disk / Linux and Verint Software 6.0 GB 6.0 GB on first disk Swap Virtual Memory Twice RAM Twice RAM on first disk Issue 5 May,

82 Installation Mount Point Use One Physical or Logical Disk /var Linux /var and the database 1-99 GB (configurable) /calls The recordings Remainder of the disk Two Physical Disks Remainder of the first disk Whole of the second disk WARNING:! WARNING: The size of /var must be calculated carefully based on the number of recordings per month and how many months the database records will be retained. Creating the kickstart script The preferred method requires a Windows machine with a CD drive and a floppy drive. The recorder must also have a (non-usb) floppy. If these requirements cannot be met follow the alternative process described under Performing a kickstart install without a floppy on page Format a floppy and insert it into the floppy drive of the Windows machine. 2. Insert the installation disk into the CD drive of the Windows machine. The kickstart script generation tool will start automatically. 3. Fill in the form with the following information: a. Select the keyboard layout of the recorder server b. Select the timezone of the recorder server c. If you have a corporate NTP server specify its IP address or fully qualified domain name. If you leave the entry empty the tool picks a suitable public server. WARNING:! WARNING: Because good time synchronisation is so vital you must take care with this setting and test that it is working after installation. d. Fill in the address and netmask for the first NIC. e. Specify the address of the default router, which must be in the same subnet as the address that you specify for the first NIC. f. Specify the hostname, preferably as a fully qualified domain name (e.g. cscm1.bigcorp.com) g. Specify the IP address of a DNS server h. If there is a second NIC and you want to enable it, fill in the address and netmask. This address must not be in the same subnet as the first NIC. 82 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

83 Platform Prerequisites i. For one logical drive or physical disk, select 1 disk, and specify the size of the /var partition. For two logical drives or physical disks, select 2 disks, and specify the device names of the two devices. If you do not know the names of the two devices, follow the first few steps of a normal interactive RedHat installation. After selecting the keyboard type, the RedHat installer will display a menu showing the names of the devices. j. Select the number of Linux tools to install. Minimal chooses just those parts of Linux needed to run the recorder (Choose this option only if you are familiar with Linux and will perform all installation from the command line.) Recommended installs other useful management tools and the Linux windowing system. 4. Click Generate Floppy. 5. Eject the floppy. Performing the kickstart install 1. Have the floppy just created ready. 2. Boot the target server using the first disk of the RedHat distribution. WARNING:! WARNING: Make sure that the disks are the Update you require. RedHat and other vendors still sometimes supply Update 0 disks. 3. Wait for the boot: prompt. 4. Insert the floppy. 5. Type linux ks=floppy 6. Wait for the automated install to complete and insert the other disks when requested. Performing a kickstart install without a floppy If it is not possible to perform the recommended floppy-based kickstart installation, this HTTP based procedure may be used instead. 1. Follow the same procedure to fill in the kickstart script generation tool, except at the last step click Generate File. Use the file chooser to specify the location where the file will be saved. Make sure the file is called ks.cfg. 2. Copy the ks.cfg to a web server and make it available as a web URL. The web server must not use a non-standard port. It must use port 80. Be certain when copying the file that it receives no text processing. If copying it using FTP be sure to use BINARY mode. Do not edit the file. Test that you can view the file using a regular browser. 3. Ensure that the target server will be able to obtain a dynamic IP address using DHCP on its first NIC. If necessary, temporarily cable its first NIC to a corporate LAN segment. Ensure that it will be able to access the web server. Issue 5 May,

84 Installation 4. Boot the target server using the first RedHat disk, and wait for the boot: prompt. 5. Type linux ks= Expert kickstart options Great care must normally be taken to ensure that the ks.cfg file is not corrupted or edited. The process of creating this file directly to a floppy and taking that floppy directly to the target server is the recommended approach. If the file must be edited to access expert options this should be performed on a Unix or Linux computer. Windows editors introduce additional control characters that prevent the script from working properly. Unfortunately the installation of Linux appears to work, but the Verint software will not install correctly. There is almost no reason to edit the automatically generated file, but details of the script options are available at S/ch-kickstart2.html DVD Drive Install and test your DVD+RW drive by writing a test file to the disk before attempting to use it with the recording system. Network Connectivity Domain Name Server (DNS) Entries Network Routes Ensure that the IP node names of all servers that make up the recording system are stored in the appropriate Domain Name Servers. Subsequent configuration can then be done by using the host name rather than having to use numeric addresses. You should ensure that valid IP paths exist between each of the servers and from servers to the AES, CLANs and all media processing resources. See Recorder Interfaces on page 199 for details of ports used. Fault Tolerant Paths If you are deploying Master/Standby pairs of Verint ContactStore for Communication Manager Servers on the same site, it is imperative that you provide two totally independent 84 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

85 Platform Prerequisites network paths between each Master and its standby. Failure to do so will leave the system vulnerable to failure should a common component in the network paths between the two fail. In such a case, both recorders will attempt to take control of all recording, with unpredictable results. To ensure this, 1. Assign IP addresses in one sub-net (call it A) for the first NIC on the Master and the first on the Standby. 2. Assign IP addresses in a different sub-net (call it B) to the second NICs on the Master and Standby 3. Connect the NICs in sub-net A to one Ethernet switch and those in sub-net B to a different switch. (If you do not have two Ethernet switches you can use a cross-over cable directly between the two NIC cards) 4. Log on to the Standby recorder and ping both IP addresses of the Master. 5. Remove the NIC cable from the first NIC (sub-net A) on the Standby server. Confirm that you can only ping the Master on its sub-net B address. Replace the cable. 6. Remove the NIC cable from the second NIC (sub-net B) on the Standby server. Confirm that you can only ping the Master on its sub-net A address. Replace the cable. 7. Log on to the Master recorder and ping both IP addresses of the Standby. 8. Remove the NIC cable from the first NIC (sub-net A) on the Master server. Confirm that you can only ping the Standby on its sub-net B address. Replace the cable. 9. Remove the NIC cable from the second NIC card (sub-net B) on the Master server. Confirm that you can only ping the Standby on its sub-net A address. Replace the cable. If any of the above tests fail, adjust your network routing and/or firewall settings until you can be sure that there are two independent paths between Master and Standby and that these are being used rather than just being theoretically available. Issue 5 May,

86 Installation Installing Verint ContactStore for Communication Manager Note: Note: If you are installing a Central Replay Server, a Standby recorder or Slave recorder, you should first read the appropriate section in Advanced Configuration on page 173. To install Verint ContactStore for Communication Manager on Redhat 4: 1. Log onto the server at the console as root. 2. Type startx to launch the windowing system. 3. Insert the installation CD. Linux should detect the CD and bring up one or two windows. Minimise any window just showing the contents of the CD. You will be asked if you wish to autorun. 4. Autorun brings up a small window showing any versions already installed. Click the install button against each package. The button will change to saying Up-to-date. 5. Reboot the server. 6. Continue the system configuration as described in Configuration on page 95. To install Verint ContactStore for Communication Manager on RedHat 5: 1. Log onto the server at the console as root. 2. Insert the installation CD. Navigate through the Computer icon to the CD Drive icon. You will see the contents of the CD. 3. Open a Terminal Window from the Applications menu. In the Terminal window type: cd /media/cdromlabel sh./autorun 4. Autorun brings up a small window showing any versions already installed. Click the Install button against each package. The button will change to saying Up-to-date. 5. Reboot the server. 6. Continue the system configuration as described in Configuration on page 95. To install Verint ContactStore for Communication Manager on RedHat 4 or 5 from the command line 1. Log onto the server as root. 86 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

87 Installing Verint ContactStore for Communication Manager 2. Mount the installation CD On RedHat 4 type: mount /mnt/cdrom On RedHat 5 type: mount -r -t iso9660 /dev/cdrom /mnt/cdrom It may be necessary to create the /mnt/cdrom mount point using the command mkdir /mnt/cdrom If the CD drive is known by a different name on your computer (e.g. cdrom1) replace all reference to cdrom with your device name. 3. Change directory to the mount point using the command cd /mnt/cdrom 4. Locate the three rpm files on the CD using the command ls For each of the RPMs (jre, cscm, cs-service) type rpm -Uvh rpmfilename 5. Reboot the server. 6. Continue the system configuration as described in Configuration on page 95. Issue 5 May,

88 Installation Installing Verint equality Balance (V5) Install and configure this application as described in Verint equality Balance (V5) Installation Guide. Then, on the Verint ContactStore for Communication Manager, complete the settings on the Port Allocations > Quality page as described in Quality Server Ports (for Selective Quality Recording) on page 130. Look on the Status > System Overview page on the Master to confirm that the link between the recorder and the Quality monitoring application is UP. If it does not connect within 1 minute, check the Alarms page and correct any problems shown there. 88 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

89 Installing Verint equality Balance (V7) Installing Verint equality Balance (V7) Install and configure this application as described in Verint equality Balance (V7) Installation Guide. Make sure you have installed the eqconnect component. Install and configure this application as described in Verint equality Balance (V7) Installation Guide. Then, If using selective quality recording On the Verint ContactStore for Communication Manager, complete the settings on the Port Allocations > Quality page as described in Quality Server Ports on page 129. Look on the Status > System Overview page on the Master to confirm that the link between the recorder and the Quality monitoring application is UP. If it does not connect within 1 minute, check the Alarms page and correct any problems shown there. If using "ContactStore Plus" style quality recording Make sure you have installed the eqconnect component on the equality Balance server. On the Verint ContactStore for Communication Manager, enter the IP address of the eqconnect service on the System Settings > Server page. Look on the Status > System Overview page on the Master to confirm that the link between the Master and the eqconnect interface is UP. If it does not connect within 1 minute, check the Alarms page and correct any problems shown there. Issue 5 May,

90 Installation Installing Viewer The installation and configuration process for this server component is quite complex and has many options. The instructions below describe a basic approach that is appropriate for most installations. They complement the Viewer Installation Guide as they highlight easily overlooked items. You should print out the guide for ease of reference as you work through the installation. Prerequisites Before starting to install Viewer, double-check that: you have installed the Windows components IIS and Message Queuing services. you are not trying to access the server using Internet Explorer 5.0 or earlier. You must use 5.5 or higher. you have either installed SQL Server 2005 Service Pack 2 OR SQL Server 2000 plus Service Pack 4 and applied the Hotfix as specified in article of the Microsoft Knowledgebase. Follow the Viewer install guide meticulously. It is critical that you set up mixed mode security and note the System Administrator (sa) password. Consider using "sa" as the password temporarily and changing it once the system is fully configured. This helps support personnel access the system to address any initial setup problems you may have. check that all the servers in the recording system exist in DNS correctly so that you can ping each by name rather than just numerical address. You are less likely to make errors entering names than numbers. Check that you can ping the Viewer server by name from the recorders and vice versa. Installing Viewer You must install this application while logged in as an administrator. Note: Note: Do not forget to install the Core components first. Drill down into the Installation\CoreInstall folder and run setup.exe. When asked for a username and password for the Viewer account, consider leaving it at the default username and setting the password temporarily to "W1tn355" and changing it once the system is fully configured. This helps support personnel access the system to address any initial setup problems you may have. 90 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

91 Installing Viewer Create an MMC Console Viewer is easier to administer if you create a custom MMC console with the three snap-ins you will need regularly. To do this: 1. Click Start > Run > mmc 2. Select Console > Add/Remove Snap-in 3. Click Add 4. Select Component Services, Services and Event Viewer and check the Local Machine checkbox. 5. Save the console as Viewer.msc to your desktop. Configuring Viewer Follow Configuring the System in the Viewer Installation Guide, noting the corresponding points below as you go. These will provide a basic working setup that you can then return to as you become familiar with the application. 1. Using the console created above, add the local Administrator user to all four roles. 2. Open a browser on the server or any other networked computer and access If asked, log in as Administrator. Note: Tip: Note: If you see errors, try adding the server to your list of Trusted Sites - it is trying to download an ActiveX control to you. Tip: You do not need to add locations at this stage. WARNING:! WARNING: You must use the server's hostname rather than dotted IP address (nor the name localhost) when accessing the url and when entering the domain name as part of the username if you are prompted for username and password. 3. Add a system - with the following settings: a. System Name = your server name b. System Type = Contact Platform Database c. Server Name = your server name d. DatabaseName = EWareCalls (Case sensitive) Issue 5 May,

92 Installation e. Location = Default Location 4. For each Verint ContactStore for Communication Manager, add a Capture Platform with the following settings: a. Name = the Verint ContactStore for Communication Manager Server's name b. Capture Platform Type = Other Recorder c. Machine Name = your Verint ContactStore for Communication Manager server name d. Location = Default Location e. Contact Platform Database = the one you added in step 3 above f. Serial Number = the serial number shown on the Settings > License page g. Port = 8080 h. URL Path=replay (not adapter/replay) 5. Follow the section "Generating a Template". Make sure you log in using the account and password you set in Installing Viewer on page 90. Add the View, CSCMAvayaCallsView, giving it the name AvayaView (no space between), Screen Name Avaya View. 6. Log back in as Administrator. 7. Create a Basic Query. Bring up the browser on again and this time select the Viewer tab and click on the New icon at the top left to start the Avaya Query Wizard. a. Step 1/8. Select the query Avaya View that you created in step 6. Click Next. b. Step 2/8. Skip this step. Click Next. c. Step 3/8. Select the following fields: -StartedAt - Duration -FirstParty - OtherParty - Agent - Service - UCID Click the box at the top right to select all fields for the details tab. Click Next. d. Step 4/8. Select the order you wish the fields to be displayed. Click Next. e. Step 5/8. Choose StartedAt, sort descending as the default sort order. Click Next. 92 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

93 Installing Viewer f. Step 6/8. Leave the languages as the default. Click Next. g. Step 7/8. Set the name to Basic Query and leave the remaining fields blank. Click Next. h. Step 8/8. Pick an icon and set 100 maximum results. Click Finish. i. "Release" the Query by selecting the bent black arrow icon at top left. j. Use the Administration tab to make this query accessible to users. Configuring Verint ContactStore for Communication Manager For each Verint ContactStore for Communication Manager Master, Standby and Slave, you will need to do the following to configure and test the connection. Do not do this until you have proved that each recorder is recording calls successfully and that these are playable from the recorder's integral search and replay page. 1. Set the Viewer server's address on the System Settings > Server administration page 2. Check that the Link to Viewer is showing on the Status > System Overview page and is UP (allow up to 1 minute for connection to be made). 3. If not, check the Alarms page and correct any problems showing there. 4. Make a new test recording. 5. Access Viewer using a browser and search for the new recording. 6. Play the recording. Issue 5 May,

94 Installation Installing Archive Do not attempt to install or configure Archive until you have proved that Viewer is working with all recorders. Follow the installation instructions in Archive Installation Guide. Because Archive relies on Viewer, there is no additional configuration necessary on the recorder itself. 94 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

95 Chapter 4: Configuration This chapter gives details of the steps to configure a Verint ContactStore for Communication Manager system. The main sections in this chapter are: Overview on page 96 Accessing the System on page 97 Licensing on page 99 Security on page 103 System Settings on page 105 System Monitoring on page 113 Port Allocation on page 114 Search and Replay on page 136 Backup/Restore on page 141 Distributing User Instructions on page 145 Issue 5 May,

96 Configuration Overview You must now configure the recording suite to suit your requirements. This section guides you through the various tasks in a logical order. You should follow its steps immediately after installation of the Verint ContactStore for Communication Manager application. 96 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

97 Accessing the System Accessing the System Before you can configure the system, you must first: access the administration web-interface via its URL accept the license terms and conditions log in URL You administer the Verint ContactStore for Communication Manager system via a web interface. 1. Open Internet Explorer and navigate to using the name of the server you wish to administer. 2. Read the entire license agreement. If you choose to accept, click I Accept. 3. Enter a username of your choice and leave the password field blank. 4. Click OK. Note: Note: The login page uses Javascript. If you see the login page but nothing happens when you click OK, your Internet Explorer settings may be blocking this. See ActiveX Control Download on page 138 for detailed instructions on this and other necessary settings. Initial User Account The application will accept any username during the first log on attempt after installation and will automatically create a local application account for you under that name and give it administrator rights. As the password of this account has not yet been set, the web application immediately directs you to a page asking you to set the password for this account. In this instance, leave the Old Password field blank and enter a new password of your choice into the two other fields. Click OK. Issue 5 May,

98 Configuration Key Points Invalid settings Before using the System Administration pages, familiarize yourself with the following key points. Any of the system's settings that are known to be invalid are shown in red. Use the information in this guide to change the settings to valid values. If you change a setting, but submit an invalid entry, a message indicates the reason that the entry is rejected and you are prompted to re-enter it. To quit without changing a parameter, click on the Close Window link. < Previous and Next > links Show All links Page at a Time At the bottom right of some pages are links which guide you through the main configuration pages in the same order that they are shown in the left-hand navigation bar. At the top of pages that show a list of entries that spans more than one page, the Show All link appears next to the page selection tags. When you click this link, all the search results are presented on a single scrollable page. If you have clicked the Show All link described above, you can return to seeing one page of entries at a time by clicking this link. Impact of changes When you change a setting, the window into which you enter the new setting explains the meaning of that setting and the consequences of changing it. Read these notes carefully. Some settings require you to restart the recorder while others may truncate current recordings. 98 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

99 Licensing Licensing Until you enter a valid License Activation Key or select the Five day timed trial license, the application will only show you the license entry screen. Terminology License Generation Key This is a three digit number that is specific to a particular server. This is shown on the license entry page. You will need it to obtain a valid License Activation key. License Activation Key This is a long (30 or more characters) string containing the serial number, server type and channel counts that you have licensed. You must obtain this key and enter it into the administration pages before your recorder will operate. Recorder Serial Number This is a unique identifier for every Verint ContactStore for Communication Manager. For Verint ContactStore for Communication Managers, this is a 6-digit number in the range to , which is allocated by Verint Systems as part of the licensing process. The serial number is issued when the license is purchased and defines the first 6 digits of the unique reference number given to each call recorded by the recorder. For example, the recorder with serial number records its first call into the following files: wav xml The serial number is encoded within every activation key issued. Once a recorder has been configured with its initial activation key, subsequent keys must have a matching serial number. Important:! Important: The 5-day license option uses serial number This temporary and non-unique serial number is the only serial number that you can subsequently override with the recorder's correct serial number, which is included in the full license key. Issue 5 May,

100 Configuration Obtaining a License Activation Key Obtain a license activation key as follows: 1. Open another browser window (on this or another PC) and ensure that pop-ups are not blocked. 2. Navigate to the Verint licensing website at 3. Enter your Username and Password. 4. Click Log in 5. Click License Activation 6. Choose the Serial Number from the drop-down list. 7. Enter the three-digit License Generation Key from the License page in the Administration application. 8. Enter the appropriate information for the end user. 9. Click Generate Key. 10. Your license activation key is: a. Displayed on the screen b. Sent to you through Activating the License 1. Return to the Enter License Key Administration page that you have open in another browser window. 2. Enter the License Key. The license activation key is not case-sensitive, and you can omit the dashes. If you use a browser on the same machine to obtain the activation key, you can copy and paste the number between the browser windows. 3. Click Enter. The page displays the licensed serial number, server type and channel capacity. Make a note of the license key and store it safely in case you need to reinstall the application on the same server - in which case you will be able to reuse the key. To reinstall on a different server, you will need a new key, because the MAC address, to which it is tied via the three digit license generation key, will be different Once you have successfully entered a license key, you will be able to access the other pages of the administration interface. 100 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

101 Licensing Adding additional licenses Follow the procedure in the previous section titled "Activating the license" for the installation of additional licenses. Each activation key you are given includes the serial number of your recorder and its license generation key, but might not contain details of all licenses. For example, if your initial purchase included a Verint ContactStore for Communication Manager server license with 20 full-time and 30 part-time recording ports, the license key is lengthy. If you then purchase 10 additional full-time recording ports, your second key may be shorter as it need not include the part-time licenses. Note: Note: Additional licenses require additional ports on the Device, Media AND Call Control API and potentially more switch components (C-LAN, VoIP resources, media processing boards). Reinstalling on the same PC If you reinstall the recorder software on a new hard disk in the same chassis, you can reuse your existing activation keys. If you reinstall the recorder from the installation kit, you must re-enter the activation keys. If you have more than one license key, re-enter the activation keys in the same order they were purchased. Otherwise, as in the previous section, "Adding additional licenses," your number of replay licenses (included in both the keys) will end up at the lower figure if the original activation key is entered last, overwriting the higher figure (30) included in the later key. WARNING:! WARNING: If you reinstall the software, you must restore the database as described in Restoring data to a new PostgreSQL database on page 142 before starting recording. Otherwise the recorder will reuse recording identifiers that have already been used. Reinstalling the Recorder on a new PC For license security, the installation is tied to the first Network Interface Card (NIC) in the server on which it is installed. To reinstall the recorder on another server, do one of the following: Move the first NIC to the new server and use your existing activation keys Issue 5 May,

102 Configuration Note the license generation key on the new server and request new activation keys as outlined in Obtaining a License Activation Key on page 100. WARNING:! WARNING: If you reinstall the software, you must restore the database as described in Restoring data to a new PostgreSQL database on page 142 before starting recording. Otherwise the recorder will reuse recording identifiers that have already been used. 102 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

103 Security Security Security of recordings is very important and is discussed at length in System Security on page 163. At this stage in the configuration of your system you should immediately create appropriate user accounts as described below. Note: Note: If you have installed Viewer or a Central Replay Server then you should only configure Administrator accounts on each recorder. Create and control end user replay rights on the replay server. Securing the System The system automatically creates an initial administrator account as you log in for the first time. If you wish to create additional administrator accounts, you should do so now. Until you do so, the only means of accessing the system is with the initial account. To create a new user account: 1. Click on Security > Users at the left of the Administration screen. 2. Click on Add User. 3. Enter the user's name (with domain name if using Windows domain accounts). 4. Check the User is an administrator box if you wish to grant this user administrator rights. Note: Note: The Comment field is for your own notes. See Search and Replay Access Rights on page 136 for an explanation of how to use the Is allowed to replay calls owned by field. Windows Authentication If you enter a simple username without a domain qualifier (for example, admin rather than mydomain\admin) then the account is administered by the recorder application. Administrators may reset this account's password and the user may change their account password and log off from the application using the administration web pages. However, if you specify a windows domain and username, the system will attempt to authenticate you via your normal Windows domain controller or WINS server(s). In this Issue 5 May,

104 Configuration case, you will not see Logoff or Change Password links on the web pages presented by the recorder. This feature is also known as "Single Sign On" (SSO) as users only have to log onto their Windows workstation, but do not need to log on to the search and replay application separately. See Windows Domain Authentication on page 165 for details of how to enable this feature. 104 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

105 System Settings System Settings You should now configure how the system makes recordings and how it interfaces with your Avaya Communication Manager. Follow the procedures below, clicking on the appropriate link at the left of the Administration screen for each section. Server This is the section where you specify information about the recorder hardware and environment. Note: Note: It is important that you review and change any of the settings that are not correctly defaulted. Optional Internal DVD+RW Drive This setting shows the path of the DVD+RW drive being used to archive calls - for example, /dev/cdrom1. If you do not have such a drive, or do not want to use it for recording archival, leave it blank. It then shows as "Not defined". Start DVD+RW volume labels with The recorder sets the volume label on all DVD+RW media it produces. The label consists of up to five characters that you can set here, followed by a media number. If you have more than one recorder, set this differently on each one so that disks produced by each recorder can be told apart. Maximum total call duration (hours) To avoid having recording channels permanently active, the recorder will reset a channel that has been continuously recording for this many hours. Set this field to a value that you can be sure will not occur for a real call - typically just longer than any one person would ever be present on a shift. Maximum recording segment duration (mins) To optimize the playback experience, the recorder cuts long recordings into segments of the designated length. A typical value for call centers is 120 minutes. This is the default. However, if your switch regularly handles longer calls, you may increase this value Issue 5 May,

106 Configuration Time between beeps (secs) If you are using the recorder to inject warning tone onto calls as they being recorded (a mandatory requirement in some countries/regions) you can specify the interval between beeps. Retain call details for (months) At some point, the size of the call details database will become either unmanageable or will fill the available disk space. Specify, in months, how long the system should retain call detail records before they are purged from the system. This ensures that the database stabilizes at a finite size. Purging is carried out at, or shortly after, 1:00 a.m. each night and does not affect recording or replay. Add VDN number as additional "owner" of calls Access to recordings is normally controlled according to the station or agent that is the subject of the recording. You may, however, choose to control access to recordings on the basis of which VDN the call was routed by. To do this, you must first set this option to Yes. Master Recorder to shadow If this recorder is licensed as a standby recorder, you must enter the two IP addresses of the master recorder. Separate the two addresses with a semicolon. Central Replay or Viewer Server If using a Central Replay or Viewer Server, set this field to the name of that server. In the case of a Central Replay Server you MUST also specify the port number (e.g. my_crs_server:8080). Once you have administered this setting, the recorder queues call details to send to the replay server. WARNING:! WARNING: If you intend to install either a Central Replay or Viewer server and wish all call recordings to be uploaded to it, set the appropriate address to a temporary, dummy value before you start recording. Do this even if the server is not yet ready. This way, all recordings' details will be queued ready for upload when the server is available. The recorder will raise an alarm daily until you install and correctly configure the server. Ignore this alarm. equality Connect Server If this recorder is to provide recordings for an equality Balance V7 system running in "ContactStore Plus" mode, set the name of the server on which the eqconnect interface is 106 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

107 System Settings running. By default, the recorder will contact it on port If you need to override this default, add a colon and the port number. For example: myeqbserver:3030. Allow Full Database Vacuum on Startup if required Once every 6 months, the recorder must perform essential database functions. It does this on startup, but only after warning you that this will happen. If you receive an alarm saying that this is required but cannot afford to have the recorder do this on its next restart, you can defer this activity by clearing this checkbox. After each restart, this checkbox is reset and you must clear it again if you want to defer the task further. You must allow the recorder to perform these tasks within one month of being alerted. URL(s) of Unify/External control port(s) to connect to Specify the IP node name of each external controller. The port number will default to 1414 but to override this add a colon then the port number. If the recorder is supporting multiple servers, list their names separated by semi-colons. When using the PDS Connector, for example, you must set the port number to When configuring a Slave recorder, it should be configured to contact the Master recorder on port Key Management Server If you intend to encrypt recordings, refer to Encrypted File Storage on page 169 and set the name of your Key Management Server here. Key Management Certificate Passphrase If you intend to encrypt recordings, refer to Encrypted File Storage on page 169 and set the passphrase for your Key Management Certificate here. SMTP Mail "From" Address To have the recorder alarms to you or your support staff, first create an account that the recorder can use to send s. Then complete this and the following four settings. Set this one to the name from which alarm messages should originate, for example, [email protected]. By default, all alarms are sent via but you can adjust this using a properties file setting as described in Properties File on page 174. This restricts the severity of alarm sent - apart from a nightly message which is always sent to let you know the system is functioning. Issue 5 May,

108 Configuration SMTP Mail Server SMTP Username SMTP Password Enter the name of the SMTP mail server on which you have established the account that the recorder will use to send messages. If you leave this blank, the system will not send messages when alarms occur - and you can then leave the remaining settings on this page blank. Leave this blank if your SMTP server allows unauthenticated sending. If it requires authentication, set the username of the SMTP account here. Leave this blank if your SMTP server allows unauthenticated sending. If it requires authentication, set the password of the SMTP account here. The password is masked when entered in this field. Send alarm/event s to Specify the address(es) to which alarm and event messages should sent. Separate multiple addresses with a semi-colon (;). The recipient can be a local system administrator, a manned help-desk and/or suppliers' support desks if you have a support agreement that includes this facility. The system sends an message each time an alarm occurs or is resolved. It also sends an once a night as a "heartbeat" to let you know it is still operating. You should investigate any failure to receive the daily heartbeat message as it could indicate that the server has failed. Note: Note: The system will batch s for up to 10 minutes to avoid flooding users' inboxes. Confirm that you are receiving s correctly after you make any changes to these settings. SNMP Read Community You can use an SNMP monitoring system such as HP OpenView to monitor Verint ContactStore for Communication Managers. To do so, set the name of the SNMP Read Community. The recorder will then respond to SNMPV1 Get messages. The MIB files for the recorder specific variables are provided on the installation CD. For security reasons, recorders: do not allow "well known" community names like private or public do not respond to SNMP Gets until a community name has been set 108 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

109 System Settings do not use the usual port of 161, but instead use SNMP Trap Destination Enter the IP address (if any) that SNMP traps should be sent to. Communication Manager This page lets you set details about the Avaya Communication Manager. The recorder uses these settings to create softphones that it can use to record, replay and monitor calls. The settings are described below in the order in which they appear on the screen. Avaya Communication Manager Name Set this to the name of the Avaya Communication Manager system you wish to record. AE Server Address(es) The IP address of the Avaya Application Enablement Server on which the Device, Media and Call Control (DMCC) API is running. If you want to specify backup servers, add a semicolon before each subsequent address. If the first address in the list is unavailable, the recorder attempts to establish a connection with the next server specified in the list. DMCC User Name DMCC Password The user name that the recorder should use to log in to the Device, Media AND Call Control API. The password that the recorder should use to log in to the Device, Media AND Call Control API. Encrypt Media Streams Check this setting to have the audio between the recorder and VoIP resources encrypted. You can still record calls on which the other party s audio is encrypted without setting this option. If you do set it, ensure that you have configured the codec set used by the recorder s softphones to support encryption. Also note the impact on server capacity. Issue 5 May,

110 Configuration IP Station Security Code All IP softphones that register with Communication Manager must provide a security code. The code you enter must match the code entered for all the stations that you created earlier on Communication Manager for the recorder to use. This field entry is masked for security purposes. Avaya CT Server(s) Enter the IP address of the AE Server that is configured to provide TSAPI (Avaya CT) services to the recorder. If you have one or more backup servers, you can specify further addresses, separating them with a semi-colon. Names are attempted in order if the first one fails. Avaya CT Service Identifier(s) Enter the service identifier of the Avaya CT service that the recorder is to use. If you have specified multiple servers, specify the service name for each, in the same order. Separate multiple service identifiers with a semi-colon. All services must support the same login ID and password. Avaya CT Service Login ID Enter the login identifier that the recorder should use when accessing Avaya CT. Avaya CT Service password Enter the password that the recorder should use when accessing Avaya CT. Extensions Assigned to recorder This line shows how many stations you have entered in the list below - and hence the number of ports available on the recorder. Service Observe Feature Access Code Enter the Feature Access Code configured in your Communication Manager that the recorder should use for Service Observe. Use the code for No-Talk Service Observe, unless you need beep tone, in which case use the Talk-Listen code. Unassigned Capacity This line shows how many ports are still available for allocation to a specific function before using up your total licensed capacity. This value only refers to the recorder's own licenses. 110 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

111 System Settings Port Range(s) It does not indicate that the Communication Manager has a corresponding number of IP softphone licenses installed. Here you enter the ranges of stations that you configured earlier for the recorder to use. Each port on the recorder corresponds to a station number on the Communication Manager. Ideally, you should choose a single, contiguous range of station numbers. However, you can choose multiple ranges. To make subsequent administration as easy as possible, assign sufficient stations to handle your entire licensed capacity, even if you do not intend to allocate all of these ports immediately. To add a port range 1. Click Add Port Range at the bottom left. 2. Enter a range of station numbers. To edit a port range, 1. Click on the Edit link in the right-hand column. 2. Change lowest or highest port number or both. To delete one or more port ranges: 1. Click the checkbox in the Select column for each station range you want to delete. 2. Click on the Delete selected port(s) link. Note: Note: You can only delete a port range if none of its stations is assigned to any of the recording modes.! Important: Important: Set these port ranges up correctly before proceeding to the subsequent pages on which you allocate these ports to the various recording and replay tasks. For a basic, small system, this is all that you need enter here. However, on larger systems you may want to click on the Advanced link to set some or all of the following: Specify which of several servers should act as standby to a particular set of ports. Enter the serial number of the server. Force the ports to register through a specific C-LAN(s) rather than allowing AES to determine this automatically. Enter the IP address of the C-LAN they should use. Issue 5 May,

112 Configuration Important:! Important: Since the C-LAN address is only used when registering a softphone on startup, if you change the C-LAN address of a softphone that has already successfully registered, it does not re-register to the new address until you reset it, the recording service, or the server. Once you have set any of these Advanced options, this setting will be shown on the list of ports. 112 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

113 System Monitoring System Monitoring The status of your recording system is shown on several web pages beneath the Status link on the Administration interface. Here you can see: overall status summary for the server and its interfaces to other components loading levels - both current and peak individual recording channel detail Having just configured the recorder's main settings, you should now review the status and address any problems highlighted on these pages or the Alarms page. Alarms and events occurring within the system are stored in its local database. You can view them via the Alarms link on the left of the Administration page. If this link is red, it indicates that there are one or more uncleared alarms. As you address the cause of an alarm, you should clear it by clicking the checkbox to the left of it and then clicking the Clear Selected Event(s) link. You should configure alerting and/or SNMP monitoring of the system as described in System Settings on page 105. The system will then be able to bring alarms to your attention promptly. Issue 5 May,

114 Configuration Port Allocation After initial setup, configuration of the recorder on a day-to-day basis largely consists of changing what is being recorded and by which ports on the recorder. Common Settings The following settings are used on more than one port allocation page and have the same meaning on each. This does not imply that all of these settings are applicable to ALL modes. Apply Beep Tone within recorder Audio format For each recording mode, you can specify whether or not the recorder injects beep tone. For a full discussion, see Beep Tone on page 37. In most modes, you can specify whether the VoIP audio is sent to and from the recorder in G.729A or in G.711. See Audio Format on page 41 for a discussion of compression formats. Stop recording if the call drops to just one other party In some modes, users call a port on the recorder in order to record a call. Because the recorder port is then itself a party to the call being recorded, the call will stay active if any one of the other parties fails to hang up. To avoid the port getting stuck in this way, you can set the recorder to hang up if other parties on the call hang up - leaving only the recorder and one other party left on the call. Ports Configured This figure shows how many ports you have allocated to this recording mode - as detailed in the table at the bottom of the page. Unassigned Capacity This shows how many more ports you could assign before using up the available channel licenses on the recorder. (It does not indicate that the Communication Manager has a corresponding number of IP softphone licenses installed.) 114 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

115 Port Allocation Standby Recorder Serial Number If you are using ESS or LSP standby configurations you can specify the serial number of the standby recorder that provides backup services for this range of ports or stations. Leave this field blank if you have no standby or a single standby recorder. It is only required for systems where multiple standby recorders each take over a part of the load during failover. Warn when free port count falls BELOW Recording owner Where a pool of ports is used as a shared resource, this setting triggers an alarm warning that a pool is running short of available capacity. The warning message identifies the pool of ports using the Comment field (or the port number range if no comment was entered). The default for this setting is zero - and as a pool can never have less than zero ports available, the warning is effectively turned off by this setting. If you set it to 1 the recorder will alarm when the free count falls below 1 - that is to zero. And so forth. See Search and Replay Access Rights on page 136 for details of how a recording's "owner" determines which user(s) can replay it. You can use this setting to override the default ownership (station or agent number). For example, if you are configuring a pool of ports for use by the human resources department, you might find it more convenient to have all recordings made on them owned by "hr" rather than several different station numbers, as would be the case if you left the default ownership in place. You can then control access to all of these recordings by assigning replay rights over "hr" to those users entitled to play them. You can enter an alphanumeric string or a number. Leading zeroes, if entered, are significant and are retained. All characters are stored in lowercase, so replay rights are applied case insensitively. Record calls that do NOT have a VDN number? This setting specifies whether calls that do not have a Vector Directory Number (VDN) are recorded. This typically includes all outgoing calls. When it appears at the top of an administration page, this rule applies to the recording mode as a whole. It can be overridden for specific stations or ports using the Advanced settings. Filter calls by VDN and/or Skill? This option lets you control whether or not to record on the basis of a call's VDN number or the skill hunt group the call was answered by. It applies ONLY to calls that DO have a VDN number. Record calls that do NOT have a VDN number determines how other calls are handled. Select one of the following options: Issue 5 May,

116 Configuration Record all calls that have a VDN number Record only those that are on specific VDNs or skill hunt groups. You can specify one or several VDNs, skill groups, or ranges. Record only those that do NOT come in on specific VDN or skill hunt groups. You can specify one or several VDNs, skill groups, or ranges.! Important: Important: Enter VDN numbers or skill hunt groups only. Not station or agent numbers. When it appears at the top of an administration page, this rule applies to the recording mode as a whole. It can be overridden for specific stations or ports using the Advanced settings. Assigning Ports Ranges Links under "Port Allocations" let you assign these ports to any of the uses that are appropriate given the server and channels licenses you have installed. You allocate these ports to specific uses in two ways: Explicit Allocation - where you specify exactly which ports should be used. This approach is used for all cases where users dial in to or conference in to the ports in order to use them. An example is a pool of On Demand recording ports. The station numbers assigned to this pool must match those you have placed in the hunt group that you are going to tell your users they should dial for a recorder port. Automatic Allocation - where you specify what you want the ports to record or how many ports are to be used. This approach is used for modes where users do not need to dial into ports and hence it does not matter which ports the recorder selects for a given task. An example is Station Bulk recording of a particular station. You tell the recorder which station you want to record but don't care which of the recorder's stations is used to service observe the "target" station so it can be recorded. Read the text in the pop-up dialogs carefully to see whether you are being asked to specify recorder ports or targets to be recorded. When assigning ports or specifying which stations or addresses are to be recorded, you can enter single numbers or ranges of numbers. For example, entering " " is much easier to enter than 100 different numbers. Typically you need to enter some contiguous ranges and some individual port numbers. A set of recorder ports is referred to as a "pool" and is configured on the appropriate page of the administration application by entering one or more port ranges. As you enter port numbers, keep the following in mind: 116 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

117 Port Allocation Entering a range Editing ranges A range can be a single number or a set of contiguously numbered ports or stations. Within a Communication Manager system, it is possible to have port or station numbers that have one or more leading zeros. It is therefore important that you enter any leading zeros. The start and end ports in any port range must have the same number of digits even if some of these are leading zeros. The lowest and highest port numbers in a range must have the same number of digits. Therefore, if you have some 4 and some 5-digit ports, you must enter these as two separate ranges. Enter a range of ports as follows (station and address ranges are entered in the same way): 1. Click the Add port range link. The Station Range dialog is displayed. - To enter a single port, type the number in the top text box. - To enter a range, type the number of the first port in the range into the top text box and the final port in the range into the second text box. 2. Add a Comment (optional). For pooled modes, you can use this field to name a range of ports and to note any hunt group number that you assigned to these ports. The text you enter appears in status reports and warning messages as labels for the specified range. For more information about pooled modes, see Using pooled port modes on page To quit without entering the port numbers, click Close Window. To enter the port numbers and keep the window open to specify additional port numbers, click Enter and Stay Open. To enter the port numbers and close the window, click Enter and Close. Port ranges are checked for consistency with other ranges and with license conditions as they are entered. If an entry is invalid, a message indicating the error is displayed; you can change the information as necessary. You can either change or delete the ranges listed. You can change the following fields without interrupting any recordings that might be active on the port(s) affected: Comment Prompt User in... Warning Level Recording Owner VDN recording rules Issue 5 May,

118 Configuration Deleting ranges Some changes require that the softphone be reset and hence active recordings will be truncated. These are: C-LAN address Standby Recorder Serial Number Any change to the number of ports in the range Any change to the port numbers in the range A change to the Codec Deletion terminates any recording activity on a deleted port. To edit a range: 1. Click the Edit link to the right of the range you want to alter. 2. Edit the range in the port entry form. 3. Click Enter. Delete one or more ranges as follows: 1. Click the checkbox in the Select column for each range that you want to delete. 2. Click Delete selected port range(s). Advanced settings To implement advanced settings on a range of ports, stations or addresses: 1. Add or edit a range as above. 2. Click the Advanced link. 3. Enter data in the Advanced fields. 4. To quit without saving the settings, click Close Window. To enter the port numbers and keep the window open to specify additional port numbers, click Enter and Stay Open. To enter the port numbers and close the window, click Enter and Close. Using pooled port modes If you use On Demand Recording, Meeting Recording, Unify/Externally Controlled (Conferenced), Phone Replay, or Live Monitor, you can manage each range of ports dedicated to a recording mode separately. For example, if you assign two ranges of ports to Meeting Recording, you can use one for English speakers and one for French speakers. You can also track usage of individual pools. You could, for example, set up two different On Demand pools, one for a particular department on one hunt group and one for 118 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

119 Port Allocation everyone else on another hunt group. You can track the status of these pools through the Status pages. For each port range assigned to a mode, you can view activity for that port range in the System Overview and Peak Activity pages. On Demand Recording Mode Setup Ports This type of recording uses a pool of ports on the recorder that you can access via one or more hunt groups. At the top of the page are the following settings: Apply Beep Tone within recorder Audio format Stop recording if the call drops to just one other party Ports Configured Unassigned Capacity All of these are explained in Common Settings on page 114. The table at the bottom of the page lets you assign specific recorder ports to this recording mode. Refer to Assigning Ports on page 116. In most cases, you will want to place each of these ports in a hunt group and make your users aware of this hunt group number and/or set up a key on their phones to access it. Advanced Settings Audix Click on the Advanced link when entering or editing a port range to set: Standby Recorder Serial Number Free port count warning level Recording owner All of these are described under Common Settings on page 114. To allow a user easy access to On Demand Recording from a station, combine On Demand Recording with the "One-Step Recording via Audix" Communication Manager Issue 5 May,

120 Configuration feature. To do this, enter the hunt group used for some On Demand ports as the parameter for the Audix-rec feature button you assign to the user's station. When using the Audix-rec feature, keep in mind: Recording beep tone can only be applied by the Communication Manager, not by the recorder The user pressing the button will not hear the beep tone. Calls recorded using this feature are only indexed with the party that pressed the button, not the other party on the call. For more information, refer to the One-Step Recording via Audix topic, Feature Related System Parameters Section, in Chapter 19: Screen Reference of the Administrator Guide for Avaya Communication Manager. Meeting Recording Mode Setup Ports This type of recording uses a pool of ports on the recorder that you can access via one or more hunt groups. At the top of the page are the following settings: Apply Beep Tone within recorder Stop recording if the call drops to just one other party Ports Configured Unassigned Capacity All of these are explained in Common Settings on page 114. Note that all Meeting recordings are performed in G.711 to make the production of custom voice prompts simpler. The table at the bottom of the page lets you assign specific recorder ports to this recording mode. Refer to Assigning Ports on page 116. In most cases, you will want to place each of these ports in a hunt group and make your users aware of this hunt group number and/or set up a key on their phones to access it. Advanced Port Settings Click on the Advanced link when entering or editing a port range to set: 120 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

121 Port Allocation Custom Prompts Standby Recorder Serial Number Free port count warning level Recording owner. Tip: Tip: On Meeting Recording ports, a user can enter a list of owners manually using the dial pad. All of the above are described under Common Settings on page 114. Prompt users in - Sets the language for spoken prompts. If using multiple languages, you should configure a hunt group to correspond with each range of ports that uses a different language. If the language you require is not offered, select Customer defined prompts and provide your own set of prompts. These files are located in /wav beneath the folder into which you installed the recorder and are called: welcome_custom.wav owners_custom.wav recording_custom.wav help_custom.wav You can replace these four files with your own recordings. Listen to the corresponding file in a language you understand (for example, welcome_english.wav) and record the equivalent messages in your own language. WARNING:! WARNING: You MUST use the same G.711 µ-law encoding that the supplied files use. Station Bulk Recording Mode Setup Unlike the previous recording modes, this type of recording dedicates a port on the Verint ContactStore for Communication Manager to each station it records. You do not specify which ports the recorder should use. You simply list the stations that you want to record. The system chooses from the available ports. At the top of the page are the following settings, most of which are are explained in Common Settings on page 114: Issue 5 May,

122 Configuration Stations Apply Beep Tone within recorder Audio format Delete Recording by entering - You can specify a digit string that the user can dial to delete the recording of the current call. When this delete command is issued, call segments from the past 24 hours with the same UCID and owner as the current call are deleted from the local database and the audio (wav) and details (xml) files are removed from the hard disk. If you enable this option, be sure to inform the users of the selected string. When choosing the string for this command, consider the following: - A very short string, such as *, can be entered by accident. - A longer string is more difficult to remember and enter. Consider using the alphanumeric equivalent for a digit key, for example, **D is really **3 but D for delete is easier to remember. To disable this option, set this parameter to an empty string - which will show as Not Defined and the feature is disabled. Viewer and Central Replay Server do not support call deletion. Since audio files are stored on the local recorder hard drive, the wav file is deleted when the command is given, but the call detail record remains on the central database. So, the deleted calls can be searched for but not replayed. Also, if any of the archive mechanisms copy the recording to an archive before it is deleted, then this copy will still be present. Record calls that do NOT have a VDN number? Filter calls by VDN and/or Skill Hunt Group? Ports Configured Unassigned Capacity The table at the bottom of the page lets you specify which stations are to be recorded. The system determines which ports will do the recording. It does this to minimize the impact of changes, leaving previously assigned ports untouched if possible to allow updates to configurations to be done during business hours rather than having to wait for close of business. Important:! Important: As the recorder picks the ports to use for this mode you must be careful when YOU choose to assign ports to the other modes. Use the Status > Port Details page to check if the ports you are about to assign are in use for this mode before doing so. You can still assign them to the other mode but any recording in progress will be truncated before being re-established on another port. For complete instructions on adding, editing and deleting stations, refer to the earlier section "Port(s) and station(s)." 122 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

123 Port Allocation WARNING:! WARNING: Do NOT enter Agent, Skill hunt group or VDN numbers. You can only record STATIONS with this recording mode. Advanced Port Settings Click on the Advanced link when entering or editing a port range to set: Standby Recorder Serial Number Recording owner Record calls that do NOT have a VDN number? Filter calls by VDN and/or Skill? All of the above are described under Common Settings on page 114. Station Executive Recording Mode Setup As with Station Bulk recording, you do not specify which ports the recorder should use. You simply list the stations that you want to record. The system chooses from the available ports. At the top of the page are the following settings, most of which are are explained in Common Settings on page 114: Apply Beep Tone within recorder - note that a third option is available in addition to the normal Yes and No. You can choose to have beep tone start only once the user has entered the retain command. Audio format Retain Recording by entering - You can specify a digit string that the user can dial to retain the recording of the current call segment. Normally, the recording is deleted by default as soon as the user ends the call segment, but entering the Retain digit string during the course of the call instructs the recorder to retain the recording. If you enable this option, ensure that you inform the users of the selected string. When choosing the string for this command, consider the following: - A very short string, such as *, can be entered by accident. - A longer string is more difficult to remember and enter. - You can use letters instead of numbers. For example, **R is actually **7 but R for retain is easier to remember. Ports Configured Issue 5 May,

124 Configuration Stations Unassigned Capacity The table at the bottom of the page lets you specify which stations are to be recorded. The system determines which ports will do the recording. It does this to minimize the impact of changes, leaving previously assigned ports untouched if possible to allow updates to configurations to be done during business hours rather than having to wait for close of business.! Important: Important: As the recorder picks the ports to use for this mode you must be careful when YOU choose to assign ports to the other modes. Use the Status > Port Details page to check if the ports you are about to assign are in use for this mode before doing so. You can still assign them to the other mode but any recording in progress will be truncated before being re-established on another port. For complete instructions on adding, editing and deleting stations, refer to the earlier section "Port(s) and station(s)." WARNING:! WARNING: Do NOT enter Agent, Skill hunt group or VDN numbers. You can only record STATIONS with this recording mode. Advanced Port Settings Click on the Advanced link when entering or editing a port range to set: Standby Recorder Serial Number Recording owner Both of the above are described under Common Settings on page 114. Delayed Call Deletion Optionally, you can configure the recorder to wait before deleting a recording made in this mode. If a user decides to retain a call after it has completed, he can do so by ringing a specific number (the "Retain Port") on the recorder. Note: Note: This feature requires one additional port on the recorder. To implement this configuration, you must add the following lines to the properties file: execmode.deletedelaymins=nn execmode.retainnumber=nnnnn 124 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

125 Port Allocation Where NN is a number of minutes from the end of the call within which the user can call the "retain" number to retain the call NNNNN is the station number of an otherwise unallocated port on the recorder that will be dedicated to receiving Retain commands after hang up. If you set these properties, call segments recorded in Station Executive mode are not deleted until the time specified has elapsed after the end of that recording segment. A retain command entered during or after the recording ends (within the specified period) will preserve the calls. When determining an appropriate value for this delay, consider the following: You want to maximize the chance of retaining all segments of a call that the Station Executive user chooses to Retain, but You should minimize the time in which unwanted or unauthorized recordings are available for replay. The Retain command applies only to the most recent call on a station. So, if a call is placed on hold and a consultation call is made, this consultation call is now the most recent. Calling the retain number when the previous call is still on hold results in the consultation call being retained. However, if the user resumes the held call, hangs up, then dials the retain number, the original (and final) call is retained. In this case, the segment before the call was placed on hold is retained, as long as it ended less than NN minutes before the retain command was given. Conferenced Recording This type of recording uses single-step conferencing to record calls. It offers a flexible alternative to Station Bulk and Unify/External control for rules-based recording. Instead of relying solely on DMCC, this mode uses the more powerful CTI capabilities of TSAPI (also known as Avaya CT), an extra cost option, to track call activity and control recording. The recorder does not dedicate a softphone to each recording target. Nor does it suffer from some of the limitations noted with respect to recordings made via the service observe feature of Communication Manager. Recording is triggered by rules that can be applied to four types of "target" addresses: Stations Agent IDs Skill hunt groups VDNs Issue 5 May,

126 Configuration Important: Important:! Important: There are a number of prerequisites (including additional Avaya licenses) and settings that are required for this recording mode.! Important: If you are using a Security Database as part of your Avaya CT setup, you must ensure that the recorder is granted access to all the addresses (stations, VDNs and skill hunt groups) that it will need to observe. Mode Setup At the top of the page are the following settings. Those that are simply listed here are explained in Common Settings on page 114. The other settings are unique to Conferenced mode. You must set ALL of these parameters in accordance with the instructions below before you specify which addresses are to be recorded. Apply Beep Tone within recorder Important:! Important: Setting this to Yes means that every Conferenced recording will use an extra timeslot on the port network handling the call. Audio format VDN(s) to Observe - If you want the recorder to tag and/or control according to the VDN(s) through which a call is routed, you must enter all of the VDNs in use here. Separate VDNs with a semi-colon. You can enter ranges of VDNs - but only if all values within the range are valid VDNs. Tip: Tip: If you have ranges of VDNs within which some numbers are not used, consider creating these as VDNs to allow you to enter the whole range here. The next two settings operate as described in Common Settings on page 114 but note that these are secondary tests on a call. Calls must match an address shown at the bottom of the screen before they are even considered for recording. These tests are then applied to filter the calls further. These settings should only be used if your main recording rules (at the bottom of the screen) target stations or agents. Record calls that do NOT have a VDN number? Filter calls by VDN and/or Skill? Test VDN filter against which VDN? - A call may be routed through several VDNs. If you have specified a rule in the previous setting, you can use this setting to tell the system whether your rule refers to the first, last or any VDN that the call has gone through. The default is to use the first VDN. Note that the recorder will only be aware of VDNs that you listed above. You therefore have considerable flexibility should you wish to ignore specific VDNs. 126 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

127 Port Allocation Addresses Tip: Tag calls with which VDN? - Each recording can only be tagged with a single VDN. You can choose whether this is the first or last VDN that a call went through. As with the previous setting, "first" and "last" are "as far as the recorder is aware" i.e. restricted to those VDNs you tell it to observe. Maximum Concurrent Recordings (on this recorder) - determines how many of this recorder's ports are assigned to this recording mode. If this recorder is controlling other Slave recorders, their ports will be in addition to this count. Agent Skill Group(s) to Observe - Avaya CT does not let the recorder observe AgentIDs directly. If you wish to record calls based on AgentID - or even to tag recordings with Agent Ids and names, you MUST configure this setting. Enter enough skill groups to ensure that each agent you wish to record is in at least one of these groups. Separate skill groups with a semi-colon. You can enter ranges of skills - but only if all values within the range are valid skill hunt groups. Tip: Consider creating one dummy skill hunt group and assign all agents to this skill. You then need only enter that one skill hunt group here. This also minimizes the number of Avaya CT licenses you need. Ports Configured - this should match the Maximum Concurrent Recordings figure that you have specified. If it does not, you should provide more recorder ports. Unassigned Capacity The table at the bottom of the page lets you specify which addresses are to be recorded. These addresses can be station numbers, agent logon ids, skill hunt groups or VDN numbers. WARNING:! WARNING: Using a mixture of different address types can give confusing results. You should normally choose which type of address you are going to use and not use the other three. The exception to this is if you can clearly partition your traffic so that there is no chance of multiple rules firing for a single call. You can target single addresses or ranges. For example, if your 45 agent identifiers are all in the range of 7400 to 7450, you can use the range , even though not all these numbers have been assigned. Use this feature sparingly though as it will take longer for the recorder to start and recover from outages if it has to try many more addresses than are strictly necessary. Issue 5 May,

128 Configuration! Important: Important: If you are targeting Agent logon IDs, the recorder notes which of these have been identified as valid agents and does not attempt to observe them directly. Instead, it waits to hear that they have been logged on. Should you ever change your numbering plan so that numbers previously used for Agent Logon IDs are now used as station numbers, you must restart the recorder if you wish to record these stations. The system automatically determines which ports will do the recording. It does this to minimize the impact of changes, leaving previously assigned ports untouched if possible to allow updates to configurations to be done during business hours rather than having to wait for close of business.! Important: Important: As the recorder picks the ports to use for this mode you must be careful when YOU choose to assign ports to the other modes. Use the Status > Port Details page to check if the ports you are about to assign are in use for this mode before doing so. You can still assign them to the other mode but any recording in progress will be truncated before being re-established on another port. You can enter more addresses than you have assigned ports to this mode. The number of ports assigned is controlled by the Maximum Concurrent Recordings (on this recorder) setting above - not by the number of entries in this table. For complete instructions on adding, editing and deleting addresses refer to the earlier section "Port(s) and station(s)." Advanced Settings Click on the Advanced link when entering or editing an address range to set the following options. Where these are simply listed below, full details can be found under Common Settings on page 114. Those settings that are specific to this mode are described in full. Standby Recorder Serial Number Recording owner Record calls that do NOT have a VDN number? Filter calls by VDN and/or Skill? The above two settings should only be used where the address range you are applying them to is an agent or station range. Trigger recording on alert - If this box is ticked, a call will be recorded if it alerted the specified address - even if it was never actually connected to this address. If the box is cleared, calls are only recorded if they actually connected to the specified address. 128 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

129 Port Allocation Note: Note: As calls are never actually connected to VDNs or Skill Hunt Groups (they alert on them but are connected to a station or agent) you must leave this box checked when specifying VDN or Skill Hunt Group address ranges. Continue recording to end of call - If checked, a call will be recorded until the call ends. If cleared, the recording will stop once the address that triggered recording is no longer connected to the call. For example, if the recorder is only configured to record calls made by Agent 1234, then if this box is cleared, the recording will end if Agent 1234 transfers the call to another agent. If the box is checked, recording will continue. As with the previous setting this must be checked if the address range specifies VDN or Skill Hunt Groups. Important: CAUTION:! Important: These last two settings can cause the recorder to make more concurrent recordings than you originally planned. As calls are diverted to or transferred off to other stations, those you have targeted in this mode can take other calls while those calls are still being recorded.! CAUTION: This setting cannot be used if any other application is using the Avaya "Take Control" feature to control these calls. In order to maintain a CTI observer on a call after it has left the originally observed address, the recorder implicitly performs this "Take Control" function itself and only one application can do so. Note that this effectively blocks the use of recording by VDN or Skill Hunt Group if another application is controlling call routing. In this case, you must specify station or agent ids as recording targets and uncheck this option. Master/Slave Configurations Because this recording mode can target skill hunt groups, agent ids and VDNs, it is not easy to partition a large site into several independent recorders (as you can with Station Bulk recording, for example). If you need to record more channels than a single server can handle, you can configure Conferenced recording on one (the Master) and provide one or more Slave recorders to provide the required number of concurrent recording channels. To configure a system in this way: 1. Configure the Conferenced recording mode on the Master as normal but specify 0 Concurrent Recording Channels (on this recorder). 2. On each Slave recorder, allocate all ports to On Demand mode and specify port 1416 on the Master recorder as the external controller. 3. On each Slave recorder, add the following line to their properties file: ondemand.ocpneeded=false Issue 5 May,

130 Configuration If using a Standby recorder (recommended to avoid the Master becoming a single point of failure in a large system), 1. Configure the standby as you would normally, specifying the Master recorder as the primary recorder it should shadow. 2. On each Slave recorder, add the standby's IP address (again port 1416) to the list of external controllers. Quality Server Ports (for Selective Quality Recording) Mode Setup The Quality Server Ports page is available if you have purchased one or more Quality Recording channel licenses. On this page, you specify the settings required for the recorder to interact with one or more equality Balance Servers and which ports on the recorder are available for it/them to use. At the top of the page are the following settings. Those that are simply listed here are explained in Common Settings on page 114. The other settings are unique to this mode. Apply Beep Tone within recorder Audio format Important: Note:! Important: If you want to use these quality ports to record stations that are also being recorded in other service observe based modes, you must use the same codec as these other modes. URL(s) of Quality server(s) to connect to - Specify the IP node name of the Quality server. The port number will default to If the recorder is supporting multiple servers, list their names separated by semi-colons. Username/Password for file-share on Quality server - Recordings made under the control of this Quality application are only held temporarily in the recorder's bulk recording call storage area. As each call completes it is copied onto a location maintained by the Quality application. If you have configured equality Balance to store recordings in a file-share, you must enter a username and password here so that the recorder can write recordings to and read recordings from the file-share on the quality server.. Note: When you use more than one Quality server, you must use the same username and password on each one's fileshare. Ports Configured Unassigned Capacity 130 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

131 Port Allocation Ports The table at the bottom of the page lets you assign specific recorder ports to this recording mode. Refer to Assigning Ports on page 116. For this mode, you typically add a single range of ports - large enough for the total number of recording and replay ports. The split between recording and replay is set up on the Quality Server not the recorder. Advanced Settings Click on the Advanced link when entering or editing a port range to set: Standby Recorder Serial Number This is described under Common Settings on page 114. Unify or Externally Controlled Ports Mode Setup! CAUTION: CAUTION: Since Version 7.7, this mode is no longer required for most external control applications. Ports in the other recording modes can now be controlled and/or tagged by external controllers such as Unify. There is therefore no need to use this mode. Existing users should consider reconfiguring their recorder to remove reliance on this mode. Allocate ports to this mode if you want them to be controlled by or have additional tagging applied by an external application. This may be Unify, a Verint supplied CTI interface (for example PDS or AIC connectors) or a third party application. At the top of the page are the following settings. Those that are simply listed here are explained in Common Settings on page 114. The other settings are unique to this mode. Note: Note: There are several additional properties file settings that can be used to modify how the recorder controls these ports. Apply Beep Tone within recorder Audio format Record Calls Using - Use this option to specify how the recorder interacts with the external controller(s). - If you select Service Observe, this mode is configured and acts like Station Bulk recording. You specify the set of stations to be recorded; the recorder will service Issue 5 May,

132 Configuration Ports or Stations observe each of them. According to a properties file setting, the recorder either starts and stops recordings or waits for the external controller to tell it to start and stop. - If you select Single-step Conference, Unify or the external server must determine what to record - either by establishing a single-step conference with a recorder port or asking the recorder to do so for it. In this case, the external server must provide most of the call information.! Important: Important: If you change this setting you must restart the recorder. The next two settings are only used if Service Observe is selected above - in which case they act in the same way as defined for Station Bulk Recording on page 121. Record calls that do NOT have a VDN number? Filter calls by VDN and/or Skill Hunt Group Ports Configured Unassigned Capacity If you have selected Record calls using Single-step Conference, the table at the bottom of the page lets you assign specific recorder ports to this recording mode. Refer to Assigning Ports on page 116. You would normally assign a single range of ports in this mode. If you have selected Service Observe, this table lets you specify the stations to be recorded - as described for Station Bulk Recording on page 121. Advanced Settings If you have selected Record calls using Single-step Conference, click on the Advanced link when entering or editing a port range to set: Standby Recorder Serial Number Free port count warning level Recording owner If you have selected Record calls using Service Observe, click on the Advanced link when entering or editing a station range to set: Standby Recorder Serial Number Recording owner. Record calls that do NOT have a VDN number? Filter calls by VDN and/or Skill Hunt Group? 132 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

133 Port Allocation All of these are described under Common Settings on page 114. Phone Replay Ports Mode Setup Ports The Phone Replay Ports page is only available if you have purchased one or more telephone replay port licenses. This mode requires no manual settings. The top of the page simply shows: Ports Configured Unassigned Capacity These are explained in Common Settings on page 114. The table at the bottom of the page lets you assign specific recorder ports to this recording mode. Refer to Assigning Ports on page 116. For this mode, you typically add a single range of ports - enough to handle the maximum concurrent number of replay users the recorder is to support. Important:! Important: Replay ports impose a significant load on the recorder. Be sure that you have specified a powerful enough server. Advanced settings Click on the Advanced link when entering or editing a port range to set: Standby Recorder Serial Number Free port count warning level These are described under Common Settings on page 114. Live Monitor Ports The Live Monitor Ports page is only available if you have purchased one or more live monitor channel licenses. Issue 5 May,

134 Configuration Mode Setup Ports At the top of the page are the following settings: Note: Audio format Note: This must match the format used to record the stations you wish to live monitor. Ports Configured Unassigned Capacity All of these are explained in Common Settings on page 114. The table at the bottom of the page lets you assign specific recorder ports to this recording mode. Refer to Assigning Ports on page 116 For this mode, you typically add a single range of ports - enough to handle the maximum concurrent number of live monitor users the recorder is to support. Advanced settings Click on the Advanced link when entering or editing a port range to set: Standby Recorder Serial Number Free port count warning level These are described under Common Settings on page 114. User Account Setup If you want to use this feature, you must setup an additional user account for each agent or station that you want to authorize to use it: The "username" is actually a number - and must match the agent number (if logged in - otherwise the station number) that the caller rings in from. (Or the Automatic Number Identification (ANI) or Calling Line Identifier (CLI) if you want to let the caller ring in from an outside number.) Give the user account replay rights over the station(s) that you want to let that phone live monitor. 134 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

135 Port Allocation Hunt Groups You might want to configure a hunt group on the switch to include all of these live monitor ports; this configuration makes them accessible through the single hunt group number on a "first come, first served" basis. Issue 5 May,

136 Configuration Search and Replay If you are deploying a centralized search and replay system using Viewer, you should see Viewer Installation Guide for instructions on how to customize, deploy and use the application. However, you should still prove that you can search for and replay calls directly from the recorder because: if you cannot, then Viewer will not be able to play calls either. in the event of server or network failure you may need to access recordings directly from the recorder. The instructions below relate to the integral search and replay application that is part of the recorder itself. The accompanying User Guide provides end user instructions on this application. As the system administrator, however, you may wish to: control which users can search and replay recordings customize the search fields and/or display ensure that the ActiveX control used can be downloaded to your clients' PCs modify default behavior via the properties file The second of these is rarely required and is described with other advanced topics in Customizing Search and Replay on page 186. The others are needed in most cases and are described below: Search and Replay Access Rights When a recording is made, the recorder assigns its "owner" or, in some cases "owners" as follows: If the Recording owner field is set on the Advanced settings for the address being recorded, the owner will be the number or name specified there. If an owner is not specified but the call was made by a logged in Agent, then the owner is that Agent's ID. If an owner is not specified and the call was not made by a logged on Agent, then the owner is the station that was recorded. If Add VDN as additional "owner" of calls is set on the System Settings > Server page, then calls routed via a VDN will have a VDN as another owner. In the case of Station Bulk recording, this is the last VDN the call went through. In the case of Conferenced Recording, it is either the first or last according to how you have configured the recording mode. 136 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

137 Search and Replay Note: Note: Conferenced Recording is slightly unusual. Where an agent is logged in, the recording will be "owned" by the station the call was taken on AND the agent who was logged on. Meeting Recording however, does not follow the above rules. In this mode, the voice prompts advise the caller to enter one or more owners. Tip: Tip: Use dummy station identifiers to allocate owners to calls made in Meeting Recording mode. All members of a particular team can be configured with replay rights for a particular number, even though this is not a valid station number. When prompted at the start of the call, mark meetings recorded for and by a team with this "owner" so that all members of the team can access the recording. You control which recordings your users can search for and replay by adding a user account for each person that needs to use the Replay page and specifying which range(s) of owner they are entitled to see. To add a user account, follow the same procedure that you used in Securing the System on page 103 but do not check the User is an Administrator box. In order to search for and replay a call, the user's replay rights must include at least one owner of that call. Each user's rights are shown on the Security > Users page and are set when adding the user account. You can change these by clicking on the Edit link next to them. The initial administrator's account is automatically given access rights to all number ranges up to 10 digits. As you add other users you must specify which ranges of owners each user is entitled to replay. The number of digits is significant. A user with replay rights over cannot replay calls made by and "owned by" agent 567 though they could play calls owned by agent In this example, you might grant the user replay rights over , Each user account is automatically entitled to replay any calls where the owner field is the same as their user name. This makes it easy to assign each station to the person who uses it by setting their username on the Advanced tab of the Port Allocation > Station Bulk entry for that station. Typical examples of how to use replay rights are: A user allowed to play calls made on his own station (1234) would be given replay rights An Agent who logs on as AgentID 5012 and is allowed to replay his own calls may be given replay rights A supervisor who logs on as AgentID 5050 and manages AgentIDs and may be given replay rights 5050, , Issue 5 May,

138 Configuration All recorded stations used by the HR staff have their Recording Owner set to HR on the Advanced settings. The Human Resources Manager, who uses station 5678 may be given replay rights HR, 5678 ActiveX Control Download The Replay page downloads two ActiveX controls, which it uses to decompress the audio for replay and export. If your default browser security settings prohibit the downloading of such controls, you need to provide a means of getting the ActiveX control to your users' desktops. Internet Explorer (IE) determines rights by putting web servers into zones and then granting those zones specific rights. To access and use the Replay page with its ActiveX controls, the recorder must reside in a zone with the following rights: ActiveX controls and plug-ins - Download signed ActiveX controls - Run ActiveX controls and plug-ins - Script ActiveX controls marked safe for scripting Scripting - Active scripting On Windows Vista you must also clear the "Enable Protected Mode" checkbox. Your Intranet zone and/or the Trusted Site zone may already have these rights. If so, you need to verify that the Verint ContactStore for Communication Manager is in one of these zones. If the recorder is not in the local Intranet Zone or is not a Trusted Site, you can add it as follows: 1. In the Control Panel, open Internet Options. 2. Click the Security tab. 3. Click the Trusted Sites icon. 4. Click Sites. 5. Uncheck the require HTTPS box (unless you are forcing users to use https) 6. Enter the URL of the Verint ContactStore for Communication Manager server and click Add. Note: Note: Internet Explorer does not recognize that a certain Fully Qualified Domain Name (FQDN) and IP address are the same; you must add the URL to the list exactly as your end users are expected to type it in the address bar. 138 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

139 Search and Replay Before you advise end users of the URL of the recorder, you should make sure that your users can access the Replay page through your network and that the ActiveX controls download successfully. To test this: 1. Create a non-administrator account and assign it some replay rights 2. From a typical client machine, enter the URL for the recorder in the form: (using the recorder's IP address or hostname - assuming you have entered it in your DNS server). If using https, replace 8080 with When prompted, enter the account's Username and a blank Password. 4. Set a new password as directed. 5. Confirm that the Replay page displays correctly and that the ActiveX control is downloaded. Tip: Tip: You may wish to copy steps 2 to 4 above, fill in your URL and send them out as instructions to your end users. Modify Default Behavior You can add or modify lines in the properties file as described in Properties File on page 174 to change the behavior of the Replay page to: limit the number of results returned save filter settings from session to session restrict access to the export functions: These settings affect all users and are only updated when the Verint ContactStore for Communication Manager Service starts. Limit results returned Add the following line to the properties file to set the maximum number of calls returned from a query: viewerx.limit=nnnn where nnnn is the maximum number of calls to return. Increasing this maximum over the default of 100 results in more CPU use and higher network traffic if users choose to view or accidentally request a large number of calls. Issue 5 May,

140 Configuration Save filter settings from session to session Normally, the entries in the search filter pane are blank when you first access it. If you wish, the application can remember the last used settings and apply these instead, This feature is controlled by a setting in the properties file: viewerx.savesettings=true Restrict access to the Export function Access to the individual and bulk call export features is controlled by two entries in the properties file: Allow administrators to export calls (Default = true) replay.export.admin Allow non-admin users to export calls (Default = false) replay.export.all By default, only Administrators can export calls. Note: Note: The replay.export.all setting overrides the replay.export.admin setting, so if replay.export.admin=false but replay.export.all=true, everyone, including administrators can use the export feature. 140 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

141 Backup/Restore Backup/Restore Due to the huge volume of new files created every day, a voice recorder is not backed up in the same way as most application servers. This section guides you through the issues around backing up the application, the call details database and the recordings. Application The recorder's configuration is stored in its database (using PostgreSQL), alongside the details of the call recordings. To preserve the configuration of the server, back up the database frequently as described below. If you have not installed other applications on the server, there is no need to backup the operating system or the recorder software. It is faster to reinstall these server components in the event of disk failure. You should therefore retain the installation media and license key that you used. Backing up the Database You can back up your recorder's database using a command line procedure. (See Local Archive on page 158 for an automated method of backing up the database if you are using DVD archive.) The procedure uses the PostgreSQL pg_dump command to extract data from the database. It must be executed while the database is running. Do not stop the Verint ContactStore for Communication Manager service or the Postgresql service before proceeding. To back up your postgres database: 1. Log on as root. 2. Become the database owner by typing su - postgres 3. Create a backup file by entering the command: pg_dump --format=c --compress=5 eware > backupfile You should specify a full path for the backupfile, and consider moving the resulting backup file to external media or another machine. Please observe the following guidelines concerning the compression factor: 5 is a modest compression factor. using a higher number (maximum is 9) makes the backup slower and uses more resources. However, it results in a smaller backup file. Issue 5 May,

142 Configuration using a smaller number makes the backup faster and uses fewer resources. However, it results in a larger backup file. Restoring data to a new PostgreSQL database Note: Note: You can only restore data to the server from which you dumped it because the dump file stores the software serial number and license key information. These are tied to a MAC address on the recorder. Unless you can move the original NIC into the new server, you will need to obtain a new license key if you wish to restore to different hardware. The following process erases the default database that exists after a complete re-installation and replaces it with the database that you have backed up. To restore the database: 1. Re-install the operating system. 2. Log on as root and install the recorder as described in Installing Verint ContactStore for Communication Manager on page Stop the Verint ContactStore for Communication Manager service. 4. Become the database owner by typing su - postgres 5. Drop the existing database by entering the following command: dropdb eware 6. Create an empty copy of the postgresql database by entering the following command: createdb eware 7. Restore the data by entering the following command: pg_restore --dbname=eware --use-set-session-authorization backupfile Backing up Voice Recordings The Verint ContactStore for Communication Manager stores voice recordings in the /calls partition. This partition quickly fills up with thousands of directories and millions of files. When the partition is nearly full, the recorder maintains only a tiny amount of free space on the partition by deleting batches of 100 recordings (and the directory that catalogued them) at a time, as it requires space for new recordings. This causes a huge churn of files every day. 142 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

143 Backup/Restore Limitations of full and incremental backup procedures On a Verint Contact Recorder server, two issues make it difficult to back up voice files: the file size the rate of change of the voice recording files Together these issues make most traditional backup strategies for the voice recordings ineffective. Traditional full backups are required more frequently than normal, which wastes backup media, and incremental backups are larger than expected because of the large churn of creations and deletions. For a backup strategy to be successful, it must be easy to restore the data if necessary. Traditional "full plus incremental" backup solutions are ineffective because these backup solutions cannot complete fully. In the event of a complete disk failure, the process restores the full backup, then the increments in chronological order. This procedure immediately overflows the disk when the restore program tries to create the increments because the partition holding the call is almost at capacity to begin with. The full plus incremental backup will fail because it runs out of disk space before it has processed the "removals" part of the procedure. Traditional restore procedures are also ineffective. If you use this solution to review a recording that has been deleted because of age, the recorder immediately deletes any restored file as part of its disk maintenance. Finally, traditional backup solutions often require locks on the disk while they work. This can seriously disrupt the working of the recorder. Two suitable strategies for audio backup DVD+RW archive This simplest and cheapest strategy is to use the built in DVD+RW archive mechanism. This is not only fully integrated with the workings of the recorder and its search and replay mechanism, but also is well suited to the incremental recording required for a recorder. As recordings are added to the calls path they are copied to DVD in an efficient manner. Even when they have been deleted from the hard disk, the recorder is still able to play them because it knows which DVD they are on and can replay directly from DVD, without an intervening 'restoration' step. Each DVD holds about 4GB, which means it can hold about 150 channel-days worth of recordings from a busy system. For less than a dollar a day, even a busy system can have limitless backup. Archive Server The second most effective strategy is to implement the Archive system. This is a rules-based system. It copies audio files from the Verint ContactStore for Communication Manager onto different, centralized disks. The data on these centralized disks is organized in a more permanent way Issue 5 May,

144 Configuration subject to less "churn" It is possible to pause the Archive Manager when required, so, if a backup process requires a disk lock, the downtime does not cause a problem with the server's operation. This pause feature, together with the way Archive organizes the audio on disk, makes this data much more appropriate for traditional full/incremental backup solutions. 144 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

145 Distributing User Instructions Distributing User Instructions Once you have configured the recorder, you should ensure that the end users know how to use it. Some users may need to know how to use and control the recording modes. Some will need to know how to search for and replay recordings. Those Using Recording or Live Monitor You will need to advise users of some or all of the following: Mode If you You should tell This information On Demand Recording Use this mode at all All potential users of On Demand recording The station and/or hunt group number(s) to dial to reach an appropriate On Demand recording port.how to use this mode. Configured any Audix-rec buttons Users of these stations How to use the Audix-rec button Meeting Recording Configured any Meeting Recording ports All potential users of Meeting recording The station and/or hunt group number(s) to dial to reach a Meeting recording port (with prompts in the appropriate language) Station Bulk Recording Enabled the delete command Users who may need to delete recordings during a call The digits to dial during a call to have a recording deleted Issue 5 May,

146 Configuration Mode If you You should tell This information Station Executive Recording Use this mode at all Users whose stations are targeted by this mode The digits to dial during a call to have a recording retained Have configured delayed retention Users who may need to retain a call after the call has ended The number of the retain port and explain that only the previous call will be retained. They must retain the call before making another call from the same station. Live Monitor Use this mode at all Supervisors that need to use it. The station and/or hunt group number(s) to dial to reach an appropriate port. The station or external phone(s) that they can dial FROM. To enter the station number that they want to monitor. Those entitled to replay calls Note: Note: This refers to the recorder's integral search and replay. If using Viewer, refer to the manuals for this product instead. The integral search and replay application is very straightforward and the online help within it is rarely needed. If you do not distribute the manual to all users you should still advise them of the following: The url (http or https) they should use to access search and replay Their username and how to log in for the first time (unless you are using Windows authentication) Any tips on which data fields would be particularly useful for them to search on (especially if you have populated any user defined fields). 146 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

147 Configuring Avaya Support Remote Access Configuring Avaya Support Remote Access To allow support by Avaya, you need to set the unique RAS server IP address assigned by the Automatic Registration Tool (ART) when you registered the system. When ART generates the unique RAS address, it sets the associated client address to the same as the server address, except with the last octet incremented by one. To set the RAS addresses: 1. Log onto the CSCM as root 2. Change directory to the /opt/witness/bin directory. 3. Enter the command: bash SetRASIP serveripaddress You do not need to enter the client IP address, because the script calculates this for you automatically. 4. Verify that the changes were made by entering the command: bash SetRASIP Issue 5 May,

148 Configuration 148 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

149 Chapter 5: Operations, Administration & Maintenance This chapter provides details of regular maintenance required for a Verint ContactStore for Communication Manager system. The main sections in this chapter are: Introduction on page 150 Status Monitoring on page 151 Local Archive on page 158 Preventative Maintenance on page 160 Issue 5 May,

150 Operations, Administration & Maintenance Introduction In addition to initial configuration, there are a number of tasks that need to be performed on an ongoing basis. This section discusses the use of the Status monitoring pages the Audit Trail preventative maintenance tasks that should be carried out on a regular basis 150 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

151 Status Monitoring Status Monitoring Status is shown over five pages that are accessed under the Status heading at the bottom left of the Administration web interface. These show the current: Alarms and Events System Overview - links to other components, local archive and storage Port States - the state of each recording channel on this recorder. Peak Activity - the peak concurrent load on the recorder (and, if it is the Master, on any Slaves it is controlling) Audit trail of all significant configuration and usage events. Alarms and Events The Alarms and Events page shows system warnings, alarms and events. The recorder stores alarms in its database. It deletes them when they are more than a month old. Check the box at the top of the page if you want to see all Alarms, including those that have been "cleared". The default is to show only those alarms that have not yet been cleared. The radio buttons at the top of the page let you select the minimum severity of alarms that are shown. The Alarms page will only show the most recent 1000 alarms (in 20 pages of 50 alarms each). If there are Informational or Warning level messages, you may need to restrict your view to Minor or Major alarms if you want to see further back in time. Use the links above or below the table to refresh the page and to clear some or all alarms. Note: Note: As long as you have set up mail account information on the System Settings > Server page, an message detailing alarms and events will be sent to the address(es) listed there. Refer to Appendix C: Alarms, for a list of alarms and events that may be generated and what to do about them. Viewing alarms and events The default on the page is to show all alarms and events that are uncleared. You can see any new or outstanding issues on first viewing the page. To change the set of events shown: 1. Click on the check box and/or radio button to specify your preferences. Issue 5 May,

152 Operations, Administration & Maintenance 2. Click on the Refresh link above or below the table. Clearing specific events New alarms and events are initially "active". To "clear" an individual alarm or event so that it no longer shows: 1. Click the check box to the left of the event. 2. Click the Clear selected events link. Clearing all events Be careful using the Clear All Events link. Clearing an alarm without fixing the problem may lead to system problems being "hidden" without your knowledge. System Overview Overall Status Links This page shows summary information about the current state of the recorder and should be checked at least daily. The title of this page includes the words: ACTIVE if the recorder is actively recording. During startup and when another recorder has taken over (e.g. a standby) this will not be present. *** NOT VIABLE *** if the recorder has identified one or more problems that prevent it from recording (e.g. disk full, cannot connect to DMCC etc.) This shows the status of each link that the recorder has with other components in the network. These include, DMCC, Avaya CT, Viewer, equality Balance (V5 or V7), Standby, Slave and Central Replay servers, The status of all these links should be UP. See Recorder Interfaces on page 199 for a comprehensive list of interfaces. Current DVD media Use this entry to confirm that the DVD drive has sensed, and is using, a valid disk. It will show one of the following values: active if it is able to record to the disk foreign if it cannot use or recognize the inserted disk 152 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

153 Status Monitoring old if a previously closed disk has been reinserted empty if there is no disk in the drive. Unless it is showing Active, or you are replaying from a previous disk, you should insert a new or the latest partially filled disk so that archiving can continue. Note: Note: The DVD entries are only shown if you have configured a DVD+RW drive on the System Settings > Server page. Current DVD Label Use this value to label the current disk. Disks are sequentially numbered. You should label the next disk (with the next available number) before inserting it. Free Space on current DVD media Use this value to determine when you will need to change the DVD media. A blank media shows 4 GB free space. Total call segments recorded to date This value shows the total number of call segments ever recorded by the system. Total call segments recorded today Use this value to confirm that recordings are being made today. If you have restarted the server today, this will show the number of call segments recorded since that restart. This includes recordings made on behalf of the Quality Monitoring application. Date of oldest call held on disk Until your disk has filled for the first time, you should monitor the available space on the drive. Check that the rate at which space is being consumed is in line with your predictions. You should be able to estimate when the disk will reach capacity and when the first calls recorded will be deleted to make way for new calls. This occurs when the free space drops below 1GB. Once the disk has started to "wrap" and calls are being deleted daily, use this figure to monitor the online retrieval capacity. If the figure starts to fall, recordings are using up your disk space more rapidly than before. The recording volumes are increasing, so you may need to expand the disk capacity before the duration of calls it can hold falls below your minimum requirement. Issue 5 May,

154 Operations, Administration & Maintenance Port Usage Summary The second table on the page shows summary information about the current state of various port pools on the recorder. The table only shows the modes that you use. The first rows of the table show the number and percentage of ports in each of the states for each mode. The final line shows the overall figure for all modes. The possible states are: Faulty These ports are incorrectly configured or have experienced an error. They will be reregistered two seconds after the initial problem, in an attempt to recover them. If the problem persists, they will back-off, doubling the time between retries until this reaches 1 minute. They attempt to reregister every minute thereafter. Starting These ports are registering or queuing to register with the Device, Media AND Call Control API. Idle These ports have registered successfully, but are on-hook and not in use. Setup These ports are off-hook and are: Attempting to establish a Service Observe session Placing or answering a call Receiving instructions from the caller, for example, with Meeting Recording or Live Monitor Important:! Important: A port used for Station Bulk or Station Executive Recording that stays in the Setup state indicates that Service Observe cannot be established on the target station. The station might not be configured to allow Service Observe or another station might already be observing it. Recording is not possible in either case. Connected These ports are in one of the following states: Have established a service observe connection but the port being observed is idle Are replay ports that have placed a call but are not currently playing a file 154 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

155 Status Monitoring Active These ports are: Recording ports that are recording a call Replay ports that are actively playing a file Live Monitor ports that are connected to a station Port States The Port States page shows the current state and configuration of each channel on the recorder: The station number of the softphone port on the recorder Its recording mode the external controller (if any) that is controlling the port whether or not a media (audio) stream is flowing to and from the port whether or not recording is enabled on this port the state of the channel (see Port Usage Summary on page 154 for an explanation of channel states) which station it is recording To reset an individual port, select the Reset link to the right of the port. This action stops any current recording on that port. To force a reset on all ports in quick succession, select the Reset All link at the upper left. The Recorded Station column for ports in On Demand and Meeting modes shows the port number that invoked the On Demand Recording and the owner of the Meeting Recording. (It shows the first one entered if multiple owners were specified at the start of the Meeting Recording.) Conferenced mode ports show multiple addresses if more than one address rule has triggered a combined recording. To update the page, click the Refresh link. Peak Activity This page shows the peak loading levels of the recorder. For each recording mode configured, it shows how many channels have been allocated the maximum number that have been active concurrently since midnight Issue 5 May,

156 Operations, Administration & Maintenance the maximum number that have been active concurrently since the date and time shown above the right-hand column To reset the monitoring period, select the Restart Peak Activity Counts link. For example, if your business has a weekly cycle, you may want to reset the monitoring period at the start of each week. Use this page to predict when to expand your recording capacity. For more detail of contention and "busy" events on the hunt groups associated with the pooled recording modes, use the Communication Manager's call detail recording tools. These tools might be useful if, for example, you are required to provide 98% availability of Meeting Recording ports. Audit Trail The Status > Audit Trail page shows administrator and user actions over a specific period. The default reporting period is the current day. You can also filter this report according to Event Type and Username. To generate a report for a different period, enter the date range in the calendar controls, and click the Refresh link. The Audit Trail functions track the following user actions: Successful user logins Failed user logins Password changes (although, for security reasons, the actual password is not stored) End user searches on the database Replay requests Live Monitor requests It also tracks all administrator actions that affect recording, such as configuration changes, manual port resets and creation or deletion of user accounts. Note: Note: Editing a station range is logged as a deletion followed by an addition. The Detail column includes the SQL statement used in searching for calls. It also uses the internal name of a setting rather than the user-friendly, localized name. This avoids any change of meaning that could occur in internationalization. Each report is restricted to a maximum of 1000 audit records. To report on more, break your reporting period into a number of smaller date ranges. Use your browser's print, save, or features to provide a permanent record of the details. To create a summary that presents all results on a single page, click the Show All link at the top. The Show All and Page at a Time links are not shown if the list of audit entries is less than one page long. 156 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

157 Status Monitoring Configuration records (which include the audit trail) are retained for 13 months. Each night after that period has elapsed, a background job deletes any records older than 13 months. If you want to retain the records longer, back up the database as described in Backing up the Database on page 141. You can change this default value of 13 months using the audit.purgemonths property as described in Properties File on page 174. Issue 5 May,

158 Operations, Administration & Maintenance Local Archive This section explains how to use a local DVD+RW drive to archive recordings, The recorder writes both the audio (WAV) and call detail (XML) files to DVD in batches as calls are recorded. Recording files are copied to the DVD disk when one of the following occurs: 14 hours of recordings have been made (equivalent to 100MB when recording in G.726 or 50MB in G.729A mode) 24 hours have passed since the last write to DVD The administration screens have an option to backup the calls details database to the end of the DVD. You should consider changing the DVD before it is completely full to leave space for the database backup. DVD+RW media control and care To find calls easily and reliably, ensure that all media are labelled, handled and stored correctly. To ensure maximum reliability and consistent high quality recording and playback: Follow the drive manufacturer's guidelines for cleaning the drive's lens. Follow the media manufacturer's guidelines for storage and handling. Check that the maximum shelf-life of the disks is adequate for your purposes and, if not, plan to copy the contents to new media within this period. Loading a disk To load a new disk: 1. Insert a blank DVD+RW disk in the drive. 2. Wait for it to spin up and be recognized. Important:! Important: Do not close the rack door as the tray will eject and automatically re-insert during the load process. 3. In the recorder Administration application, click on the System Overview link to confirm that the recorder has recognized the disk. 158 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

159 Local Archive The Current DVD media field shows that the disk has been recognized and is being written to; it also shows the volume label given to the new disk. The Free Space on current DVD media line should show 4GB or more available on a clean disk. Changing disks The System Overview page shows when the disk is full and an alarm is raised. To eject a disk that is still being recorded in order to retrieve calls from another disk, 1. Click on the Eject link on the System Overview page. 2. Insert the required archive disk. 3. When you have finished retrieving calls, replace the partially full disk so recording can continue. Important: Tip:! Important: The archival process never writes to disks out of sequence. For example, if you eject a partially full disk, then insert a blank disk onto which the next set of calls is then written, you cannot then reinsert the previous disk. Tip: If you have enough disk space left when you change from a nearly full disk to a new disk, you can automatically backup the database to the end of the disk using the Backup database and Eject link Labeling a disk As you eject each disk, label the disk itself using an approved indelible marker. Your label should indicate: The identifier of the recorder producing the disk. The identifier is Calls by default; you can change this value on the System Settings > Server page of the System Administration application. The sequential serial number of the disk. The date/time the disk became full. Issue 5 May,

160 Operations, Administration & Maintenance Preventative Maintenance This section highlights a number of administrative tasks that should be performed on a regular basis to ensure the system continues to operate smoothly. Daily Alarms Disk capacity System Status Unless you have fully automated alerting of these conditions, you should carry out the following procedures at the start of each day: Check the Alarms page for new problems. Check the available disk space. The disk where recordings are stored will appear to be at or near capacity. However, the system consistently maintains a level of 1 GB of free space by deleting older files. This maximizes the number of recordings that are available online to you. The Verint ContactStore for Communication Manager's disk manager thread deletes files on a FIFO (First In First Out) basis. Check the contents of the log files as described in Troubleshooting on page 205 and examine any errors logged since the previous check. Look at all error and warning messages, not just those generated by the Verint ContactStore for Communication Manager services. It is difficult to detect some problems automatically. Check the system status regularly via the Status > System Overview page and verify that all figures are in line with expectations as described in System Overview on page 152. Confirm channel status Use the Status > Port States page of the Administration application to confirm that the recording ports are in the appropriate states. Confirm recording and replay To confirm recording and replay: Verify that calls are being uploaded into the database. 160 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

161 Preventative Maintenance Use the Replay page to select the most recent calls to verify that calls are accessible. Confirm that the start time of these calls matches expectations. Verify that the start time corresponds to the most recent calls made on the extensions being recorded. Confirm that these calls are playable and that audio quality is good. Archive If using DVD+RW archive, check the current disk's available capacity. Change the disk when it fills. Weekly As you become comfortable with the normal operation of your recorder, you can reduce the frequency of the daily tasks. For example, if you know that the rate at which your disk is filling is not going to fill the available space for several months, you can check it weekly. Perform the following tasks each week: Disk capacity: main recording store When your recorder is first installed, the disk is almost empty. As it gradually fills, you should note the rate at which it is being used (at least weekly) and extrapolate to estimate when the disk will be full. At this point, the Verint ContactStore for Communication Manager will begin deleting the oldest calls to make room for new ones. If this happens to calls that are younger than planned, check the configuration of the recorder to ensure that only the anticipated calls only are being recorded. Add additional disk capacity to the partition before it fills. Disk capacity: other partitions Check the available space on any other disk partitions. Verify that these other drives have sufficient space. The recorder will warn you if they fall below 500MB of free space. Accumulated temporary files or log files can account for this drop in available space. You may need to purge them manually. Important:! Important: When you are purging files, remember that files you delete go to the Recycle Bin and that the space they occupy is not freed until you empty it. Call detail database purging If you have enabled automatic purging of aged call detail records, you should still monitor the size of the calls database during the first few months of use. You can then predict how Issue 5 May,

162 Operations, Administration & Maintenance large the database will get by the time old records begin to be purged. Many customers plan never to purge call detail records, but choose instead to add disk capacity every year or two as the database grows. If you do this, you should upgrade your server every few years to compensate for the increasing size of the database and the reduction in search and update speed. Configuration Backup Changes to system configuration that affect user access rights are stored in the PostgreSQL database. This means that the system configuration is backed up whenever the call detail records are. See Backing up the Database on page 141. Monthly Loading trends Check the following aspects of the system on a monthly basis: Note the total call volumes recorded every month to be aware of gradually increasing traffic trends. To do this: Note the number of calls recorded at the end of each month and compare with previous month's accumulated total. Note the age of the oldest call on the disk (only applicable once the disk has filled for the first time) Note the CPU load during busy hour If it appears that the load is increasing, consider purchasing extra licenses if required and/or increasing server specification or disk space. Every Six Months The recorder must perform a full vacuum of the database approximately once every six months. As this interval is reached, the recorder issues a daily warning message. This tells you that it will do a full vacuum on next restart - unless you postpone it by clearing the checkbox on the System Settings > Server page. You must restart the recorder and allow it to perform this database maintenance task within one month of being warned about it. 162 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

163 Chapter 6: System Security Security of customer recordings is very important. This Chapter discusses the various features - some optional - that you can use to ensure the safety and integrity of recordings. This chapter assumes that you have suitable firewall, antivirus software and physical access procedures in place. The mains sections in this chapter are: Use of SSL on page 164 Windows Domain Authentication on page 165 Blocking Replay from the Recorder on page 166 Changing Passwords on page 167 Encrypted File Storage on page 169 PCI Compliance on page 170 Issue 5 May,

164 System Security Use of SSL You should consider whether you wish to enforce the use of Secure Sockets Layer (SSL). By default, users can access the recorder via http (on port 8080) or by encrypted https (on port 8443). You can force users to use the secure https port, by setting Allow unencrypted (http) access? to No on the Security > Users administration page. When you do this, any user who attempts to access the recorder through the unsecured (http) route is automatically redirected to the secure (https) address.! WARNING: WARNING: You should not force the use of https if you use Central Replay Server or Telephone Replay. The application is distributed with an SSL certificate that is valid for 3 years from the date it was issued. The certificate makes it possible to give users secure access to the server. When users access it through this secure https port, the traffic between their browser and the recorder is automatically encrypted. However, Internet Explorer will warn your users that the name on the certificate does not match the name of the server using it. You can either advise your users that this is acceptable and should be ignored or, for greater security, you may acquire and install your own SSL certificate as explained in Installing a Signed SSL Certificate on page Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

165 Windows Domain Authentication Windows Domain Authentication You can create local user accounts within the recorder application itself. However, it is more secure to use Windows domain accounts and you may wish to enable this feature - or even restrict access so that only windows domain accounts have access to the system. To ENABLE Windows Domain Authentication: 1. Create a user account (as described in Securing the System on page 103) who's username is domain\username - for example, CORP1\JSmith. Note that the username is case sensitive and must match exactly the case of the username stored in the domain controller. 2. Add properties to the properties file to define either your domain controller or WINS controller as follows: sso.dc=ip address of domain controller OR sso.domain=domain name to use sso.wins=ip address of WINS server to use To ENFORCE Windows Domain Authentication only: 1. Enable Windows Domain Authentication as above. 2. Log in as an Administrator using a domain account 3. On the Security > Users page, set Allow local user accounts? to No. Tip: Tip: If users are prompted for their domain passwords when they access the web interface, make sure that the recorder is either part of the intranet zone, or make it a trusted site and configure Internet Explorer to automatically log on to trusted sites. Issue 5 May,

166 System Security Blocking Replay from the Recorder If you are using Viewer to replay calls, you may want to block users from replaying calls directly from the recorder itself. To do this: 1. Log in as an Administrator 2. On the Security > Users page, set Allow search and replay from this server? to No. 166 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

167 Changing Passwords Changing Passwords The recorder and related applications use a number of user account settings that are installed with a hard-coded default. You can change these as follows: Linux User Accounts The root and witness user accounts are installed with default passwords. Change these by logging on to the server and typing passwd. Be sure to note the new passwords securely as remote support staff will need these to maintain your system. Should you lose these passwords, your system will become completely unmaintainable. Postgres Database Owner You may wish to change the password used by the recorder to access its local postgres database. Before doing so, you must first enable 256 bit encryption as described in Installing Unlimited Strength Encryption on page 302 and obtain the Verint encryption tool as described in Encrypting Properties File entries on page Log in as root 2. Switch to the postgres user account by typing su - postgres 3. Access the database by entering psql 4. Change the password by entering alter user eware with encrypted password newpassword ; 5. Quit the database by entering \q 6. Encrypt the new password using the WitsBSUserCredentials tool. 7. Add the new password (in encrypted form) to the properties file as db.password=encryptednewpassword Issue 5 May,

168 System Security Recorder Account on eware/viewer SQL Server If you change the username and/or password on the eware database's SQL server, you must also set these on the recorder so that it can upload recordings into that database. To do this: 1. Log in as an Administrator 2. Enter the following url 3. Enter the username and password on the web page shown. 168 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

169 Encrypted File Storage Encrypted File Storage Recordings (WAV files) and their associated XML data files are not normally encrypted but can be - using the AES256 algorithm and RSA Security's Enterprise Key Manager. Encrypted files are also "fingerprinted" to avoid tampering. See the manual Impact 360, Full-time Recorder, Enterprise Security Administration Guide Release 7.8 for details of how to deploy an RSA Key Management Server (KMS). To then enable encryption on the recorder: 1. Install the Unlimited Strength policy files as described in Installing Unlimited Strength Encryption on page Create a KMS client certificate (a PKCS#12 file) for the recorder. This will be protected by an passphrase to preserve the confidentiality of the private key. 3. Install this certificate on the recorder by logging in as witness and copying the file to the /opt/witness/keystore directory. 4. Back at the KMS, convert the client certificate into a PEM file. The PEM file must be imported into the KMS to identify the recorder as a valid client. To convert the PKCS#12 file to PEM format use the command: OpenSSL pkcs12 -in svr_cert_key.p12 -out svr_cert_key.pem 5. On the System Settings > Server page of the recorder, enter the IP address of the Key Management Server and the passphrase for the certificate. You must restart the recorder after making any changes to the KMS settings, certificate or passphrase. Note: Note: You must also configure Viewer and Archive (if used) as described in Impact 360, Full-time Recorder, Enterprise Security Administration Guide Release 7.8 Issue 5 May,

170 System Security PCI Compliance To make your system compliant with PCI recommendations, you should adopt all of the above features. In addition: 1. Do not store sensitive data in the User Defined Fields of recordings. 2. Use the PAUSE and RESUME recorder control features to avoid recording sensitive information. This will require integration with your other systems. 3. Do not reduce the default 13 months for which audit records are kept. 4. Use encrypted audio to and from Avaya Communication Manager (setting on System Settings > Communication Manager page). 5. Force users to use Viewer rather than replaying calls from the recorder. 6. If using Quality recording, use Version 7.8 or higher. Version 5.x does not support encryption. 7. Ensure all other components of your system (Viewer, Archive etc.) are configured in accordance with the Enterprise Security Administration Guide. 8. PCI dictates a session timeout of 15 minutes. When using Windows Domain Authentication (as required to meet other PCI requirements) the recorder will accept a valid Windows logon (on an appropriate account) as sufficient to gain access to the recorder's administration pages. You should therefore ensure that all users' PCs are configured to launch a screen saver after 15 minutes of idle time and have Password Protection on Resume enabled. To ensure this, domain administrators should lock down these settings in the group policies of the domain controller. 9. Review and adjust your Windows Domain policies for user accounts if required. For instance, the PCI specification mandates the following rules: Immediately revoke access for terminated users. Remove inactive user accounts every 90 days. Do not use group, shared or generic accounts and passwords. Change user passwords at least every 90 days. Passwords should be at least 7 characters long and should include both numeric and alphabetic characters. Do not allow individuals to submit a new password that is the same as any of the last 6 they have used. Lock out the user account after not more than six unsuccessful access attempts (have a lockout duration of 30 minutes or until admin enables the account). 10. To ensure no recorded data is stored anywhere in an unencrypted format, including on the supervisor's PCs, Internet Explorer's Advanced Security Do not save encrypted pages to disk and Empty Temporary Internet Files folder when browser is closed 170 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

171 PCI Compliance settings on the supervisors' PC must be enabled and locked down. (Internet Options > Advanced > Security). 11. To ensure proper authentication when SSL is enabled in a recording system, the following advanced Internet Explorer security settings on Supervisor PCs must be enabled Check for the publisher's certificate revocation. Check for server certificate revocation. Warn about invalid site certificates. Use SSL 2.0. Use SSL 3.0. Use TLS 1.0 Issue 5 May,

172 System Security 172 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

173 Chapter 7: Advanced Configuration This chapter provides an overview of the more complex and rarely used options for a Verint ContactStore for Communication Manager system. The main sections in this chapter are: Properties File on page 174 Standby Server on page 182 Central Replay Server on page 183 User Defined Fields on page 184 Customizing Search and Replay on page 186 Usage Report on page 191 Selective Record Barring on page 193 Issue 5 May,

174 Advanced Configuration Properties File A number of system settings can be changed from their default values by placing entries in the properties file as described below: this is a plain text file, located in the installation path and with filename: /opt/witness/properties/cscm.properties you should edit this file using the vi text editor when logged on as witness - NOT root. the Verint ContactStore for Communication Manager service reads this file as it starts, so any changes made to the file will not take effect until you next restart the service. the file is normally empty as all settings default as shown in the table below. Most of these entries are discussed elsewhere in this manual, in the appropriate context. The table below provides a summary of the available settings. Entry (case sensitive) Default Meaning audioserver.inactvitytimeout 30 The number of seconds of inactivity before an audioserver port is released back to the pool of available ports. audioserver.viewerbase64 not disclosed Use this setting to specify the base64 encoded name that should be included in requests to Audio Server. You only need to set this if you have changed Audio Server from its default. audit.purgemonths 13 The number of months to keep audit trail entries in the database. conf.hanguponmediastop false Conferenced mode recording ports no longer hang up (by default) when the media flowing to them stops. If you are experiencing "stuck" or "zombie" ports left up long after calls should have finished, consider setting this. However, it will make the system more sensitive to shuiffling and may impact other applications that are watching the CTI events for these calls. 174 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

175 Properties File Entry (case sensitive) Default Meaning conferenced.defaultrecord true Whether or not conferenced mode ports will start to record as soon as they have been conferenced into a call. Rather than setting this false, consider using On Demand ports and having the external controller determine which port is used. (It is difficult to know that a conference mode port has connected to a call unless it sends a STARTED message - and it will not do this if it does not start to record when it connects to the call.) conferenced.ocpneeded true Whether or not Conferenced recordings should detect other call parties using the conference display feature. Set false if an external controller is to provide this information instead. cmapi.timeout 65 How long (in seconds) the recorder will tolerate failure of DMCC. cscm.disablecompress false Normally, recordings made in G.711 will be compressed to G kbps. Set this to true to disable compression. cscm.diskmanager true Whether to delete oldest calls when needed to create space for new recordings. cscm.localport 8080 The http port used by the recorder. Must match the entry in server.xml. cscm.logkeepdays 30 Number of days log files to retain before purging. disk.stopatmb 10 Stop writing to disk if free space is less than this many MB disk.warnatmb 500 Raise warning if free disk space falls below this many MB dmcc.port 4722 The port number on the AES for DMCC. dmcc.secure true Whether to connect to DMCC using SSL. dmcc.trustall false Allows different SSL certificates to be installed on the AES. dnis.fieldname spare2 The field name to tag DNIS information under. Applies to Conferenced mode recording only. Issue 5 May,

176 Advanced Configuration Entry (case sensitive) Default Meaning .minalarmlevel 0 Alarm level at or above which, s will be sent. Default, 0 is INFO level. Set to 1, 2 or 3 to restrict s to Warnings, Minor and Major Alarms respectively. eqc.noagents false To use equality Balance 7.x when agents do not log in, set this property to true. The recorder will then spoof an agent login on the first call taken on a recorded device. eqconnect.agentdevice Avaya Agent Device string sent to eqconnect. eqconnect.connecttimeout 120 Seconds to keep trying to connect to eqconnect on startup before reporting failure. eqconnect.heartbeatinterval 2 Seconds between sending heartbeats to eqconnect eqconnect.inactivitytimeout 10 Seconds of inactivity after which to report eqconnect failure. eqconnect.systemdevice Avaya System Device string sent to eqconnect eware.forcesameserver false Normally, the recorder looks in the SystemSQLServerView in the eware Site database to determine which SQL server to use for its calls database. When using SQL server clusters, you may need to set this property to true. The recorder then assumes that the Calls database is on the same server as the Site database. eware.recorderhostname no default value Name by which the recorder is entered into Viewer rather than its fully qualified IP name. Should not be needed after Viewer eware.sqlserverport 1433 IP port to use when connecting to SQL server. execmode.deletedelaymins 0 Number of minutes to wait before deleting Station Executive mode recordings. execmode.retainnumber no default value The station number of the recorder port to be used as the Station Executive "Retain" port. 176 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

177 Properties File Entry (case sensitive) Default Meaning jtapi.required true Normally, if JTAPI (over Avaya CT) has been configured and is in use, it is considered a critical error if it fails. If a standby recorder is present it will be asked to take over. This can be stopped if JTAPI is not critical to the recorder. jtapi.retries 3 Number of times to keep trying to establish single-step conference. jtapi.retrydelayms 500 Interval in milliseconds between single-step conference attempts. jtapi.timeout 65 How long (in seconds) the recorder will tolerate JTAPI failure to respond to heartbeats. master.localport 1416 TCP port on which the Master recorder listens for slaves meeting.defaultrecord true Whether or not Meeting recording ports will start recording automatically at the start of a call or wait for a START command from an external controller. meeting.ocpneeded true Whether or not Meeting recordings should detect other call parties using the conference display feature. Set false if an external controller is to provide this information instead. numberplan.maxlength 7 Number of digits above which a telephone number is assumed to be an external number. Used in Conferenced mode only. ocp.alwaysrefresh false In recording modes where the conference display button is used to determine who is on the call, this process will be restarted if the display indicates that a party has joined the call. In some cases, however, an existing party may be updated e.g. a trunk name/number may be replaced by a (late arriving) ANI. Set this property true to have ANY change to the display trigger a refresh of all parties on the call. Issue 5 May,

178 Advanced Configuration Entry (case sensitive) Default Meaning ocp.strictparsing false When forwarding via coverage answer groups, the conference display may not contain a single numeric field in the end of the display. Previously this would lead to the recorder continually pressed the Conf Disp button - resulting in the party being removed from the call. Now default is not to check for spaces or valid number in this string. To restore previous behaviour if this causes problems, set this property = true. ondemand.defaultrecord true Whether or not On Demand recording ports will start recording automatically at the start of a call or wait for a START command from an external controller. ondemand.ocpneeded true Whether or not On Demand recordings should detect other call parties using the conference display feature. Set false if an external controller (such as a Master, if this is a slave recorder) is to provide this information instead. ondemand.ocpblocked false If enabled, prevents OCP. Must be enabled for slave recorders. origucid.fieldname spare3 The field name to tag original call UCID information under (i.e. the UCID of the call that a consult call relates to). Applies to Conferenced mode recording only. postgres.password not disclosed Encrypted password to use for database vacuum. primary.localport 1414 TCP port on which the recorder listens for standby recorders. purge.hourofday 1 Hour at which database and log file purges will take place. Default is 1am. Valid range 0-23 qcm.keepwav false Whether to retain Quality Monitoring recordings in the recorder's call path. queue.xxx.threads 1 Number of threads to use for job queue xxx. 178 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

179 Properties File Entry (case sensitive) Default Meaning rec.keepnullucids 5 Whisper announcements are recorded in service observe modes. These result in (largely useless) recordings of a few seconds duration that can be identified as such by the lack of a Universal Call ID.(UCID) The recorder therefore deletes recordings without a UCID that are shorter than the number of seconds specified here. To turn the feature OFF, set the value to 0. rec.maskallowed false You must set this to true to enable the "pause/resume" features of the Recorder Control Protocol. rec.mincallduration 250 Calls shorter than this many milliseconds are deleted. replay.export.admin true Whether administrators are allowed to export recordings. replay.export.all false Whether all users are allowed to export recordings. (If set true, overrides replay.export.admin=false) rtp.packetlog false If true, log details of every RTP packet received slave.inactivitytimeout 10 How long the master will tolerate lack of heartbeat with slave recorder before reporting error. snmp.port 2161 Port on which recorder listens for SNMP sso.dc sso.domain sso.wins no default value no default value no default value The IP address of the domain controller to use for Single Sign On. (Incompatible with sso.domain.) The domain name to use for Single Sign On. (Incompatible with sso.dc.) The IP address of a WINS server to use to locate domain controllers. (Mandatory with sso.domain.) standby.backoffdetect true Whether the Master recorder checks for a standby taking over. standby.connecttimeout 120 Seconds a recorder will wait on startup before reporting that a previously connected primary/standby has failed to connect. Issue 5 May,

180 Advanced Configuration Entry (case sensitive) Default Meaning standby.inactivitytimeout 60 Seconds the Master will wait between polling messages before reporting a Standby as having failed. standby.localconfig false Whether the standby is locally configured. If not, it copies config from the Master. standby.reconnecttimeout 10 Seconds the standby will attempt to reconnect before taking over. standby.sscdelay 5 Number of seconds standby waits for Master to act on single step conferences before taking over. stnbulk.defaultrecord true Whether or not Station Bulk recording ports will start recording automatically at the start of a call or wait for a START command from an external controller. stnbulk.ocpneeded true Whether or not Station Bulk recordings should detect other call parties using the conference display feature. Set false if an external controller is to provide this information instead. stnexec.defaultrecord true Whether or not Station Executive recording ports will start recording automatically at the start of a call or wait for a START command from an external controller.. stnexec.ocpneeded true Whether or not Station Exedcutive recordings should detect other call parties using the conference display feature. Set false if an external controller is to provide this information instead. unify.defaultrecord false Whether Unify/external controlled ports will record by default or only when instructed to by the controller. unify.ocpneeded false Whether Unify/external controlled ports will attempt to provide call details via DMCC. usage.reporting false Whether to track actual usage to the ConfigHistory table and show the Usage report unity.required false Whether a link to Unify is required for the recorder to be viable. 180 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

181 Properties File Entry (case sensitive) Default Meaning uui.fieldname no default value User defined field name to store Avaya User to User data. If NULL, this data is not stored. Applies to Conferenced mode recording only. vacuum.interval 180 Number of days between full database vacuum. vdn.mapsize How many entries to hold in the UCID/VDN map. The recorder holds two maps and alternates between them. viewerx.limit 100 Maximum number of recordings returned for each search viewerx.savesettings false Whether to save and reuse users previous entries in search filter pane. Issue 5 May,

182 Advanced Configuration Standby Server First, determine the number and type of standby servers required using the Topology tables in Standby Recorder Options on page 254 Standby servers are installed in the same way as a standalone recorder (seeinstalling Verint ContactStore for Communication Manager on page 86 - steps 1-7). The license key determines that the server will act as a standby. Follow the configuration guidance given in the appropriate table: Where a Standby is to copy the configuration of the Master recorder (the default) then you need to configure very few other details. Work through the Configuration chapter of this manual configuring just those fields with "Edit" links next to them. Where specific standby recorders need to be specified, ensure you have set the Standby Recorder Serial Number for each set of softphones and each set of stations on the Station Bulk or Conferenced Mode settings page. Where a standby recorder is to be locally configured, 1. Add the following line to the properties file (See Properties File on page 174 for detailed instructions) standby.localconfig=true 2. Restart the Verint ContactStore for Communication Manager service 3. Configure all settings as you would on a standalone recorder. 182 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

183 Central Replay Server Central Replay Server You can deploy the recorder application onto an additional server which collects details of recordings made on one or more other servers and therefore can act as a dedicated search and replay server. Users can then search for recordings from any of the recorders feeding it with a single url and a single search. There are differences in the functionality between the Viewer application and the recorder's integral replay application as used when a recorder is deployed as a Central Replay Server. Refer to the Viewer documentation to compare it with the functionality described in the User manual for this recorder. Installation Install a Central Replay server in the same way as the Master recorder (see Installing Verint ContactStore for Communication Manager on page 86 - steps 1-7). The license key determines that the server will act as a Central Replay Server. Configuration Work through Chapter 4: Configuration configuring just those fields with "Edit" links next to them. Then configure the user accounts and replay rights for those entitled to use the application. Issue 5 May,

184 Advanced Configuration User Defined Fields The recorder can store any number of user-defined fields (UDFs) in its internal PostgreSQL database. You can search on these fields using the Search and Replay web application. Storing UDFs Use an external controller to TAG calls using the interface described in Appendix D: External Control Interface. The database automatically notes new UDFs and stores the call's details. Search and Display UDFs To search on and display UDFs, copy the browse query (named "browse") in the postgres database to a new file that you will use to create a new query. Name this "browse2" or something more descriptive. Use the stored procedure udf_value(inum, udfid) to add the chosen UDFs to the new view. For example, the following fragment of SQL inserted into the SELECT statement's list of fields adds an additional column called spare1: udf_value(calls.inum, 1) AS spare1 The udfid is determined by searching the udfnames table for the relevant udfname. The spare1 column is then available to be used in a customized search and replay view as described in Customizing Search and Replay on page 186. UDF Support in Viewer When one compares UDF support in this recorder (and hence Central Replay Server) to UDF support in eware, this recorder has the following advantages: Unlimited number of user defined fields. Unlimited length of each user-defined field. User defined search forms created by editing the xml templates rather than the end-user interface available within Viewer. 184 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

185 User Defined Fields Better support for sparsely populated data sets in which there are more user-defined fields, but not every call has every field present. Viewer Compatibility If you name your user defined fields spare1 through spare8 you can upload them to a Viewer database as well as have them in the local database. The name by which the users see these fields can be set in the template files. Issue 5 May,

186 Advanced Configuration Customizing Search and Replay By renaming and modifying the layout files provided with the Search and Replay application, you can define search forms and results display list formats. You can also provide alternate language translations for the fields and columns shown. Default Configuration Files The default layout is determined by two files located /opt/witness/properties/layouts: lcscm.xml is the default layout idefault.xml is the default internationalization file WARNING: Note:! WARNING: lcscm.xml and idefault.xml must not be modified. They are overwritten during upgrades. You will make your changes in files with different filenames (of your choice). Note: The web server picks up changes that result from additional files being placed in this folder within a minute. There is no need to restart. However, it makes sense to make these changes when there are no users on the system. Custom Layout Files Layout files are very straightforward. Each file specifies a single "layout". The filename must be of the form: lsomething.xml. The default, lcscm.xml is shown below: 186 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

187 Customizing Search and Replay <?xml version="1.0" encoding="utf-8"?> <layout name="cscm" display="ldefault"> <view>browse</view> <filters> <filter type="date" dbname="startedat" display="fdate"/> <filter type="string" dbname="partynames" display="fparty"/> <filter type="string" dbname="partyagentnames" display="fagent"/> <filter type="duration" dbname="duration" display="fduration"/> <filter type="string" dbname="vdn" display="fdnis"/> <filter type="ucid" dbname="callid" display="fcallid"/> <filter type="callset" dbname="csnames" display="fcallset"/> </filters> <columns> <column dbname="startedat" columnwidth="" formattype="date" display="cdate"/> <column dbname="duration" columnwidth="" formattype="time" display="cduration"/> <column dbname="partyagentnames" columnwidth="" formattype="" display="cagent"/> <column dbname="partynames" columnwidth="" formattype="" display="cparty"/> <column dbname="vdn" columnwidth="" formattype="" display="cvdn"/> <column dbname="callid" columnwidth="" formattype="ucid" display="ccallid" enablelikesearch="true"/> </columns> </layout> The outer tag is <layout>, within which the internal name of the layout is given by the name parameter and the display name token is given by the display parameter. It makes sense to name the file linternalname.xml but this is not mandatory. The web application presents the layouts in alphabetical order by internal name in the layout name drop list box. There are 3 inner tags: View: The name of the query on the database Filters: The filters for the left panel Columns: The columns for the grid display Issue 5 May,

188 Advanced Configuration Inside the outer filters tag, there are many filter tags. Each must have the following: a type, which can be date, string, duration, ucid or callset a column name matching one of those in the specified database view a display token which is looked up in the internationalization file. Inside the outer columns tag there are many column tags. These must have a column name, width, format and display. They may optionally have an enablelikesearch flag. See the default above for all valid display types. Widths may be valid HTML table widths (that is, nnnpx for pixels or mm% for percentages). However, it is better to leave these blank and let the browser decide. If enablelikesearch is true the application displays this field as a hyperlink that, when clicked, executes a search for all calls (+/- one day) which have exactly the same data in this column as the call clicked upon. (The example use of this is for universal call id - which you can click on to find all recorded segments of this call). Internationalization The file idefault.xml contains translation strings that allow you to format the output in any language. The current default is shown below. 188 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

189 Customizing Search and Replay <?xml version="1.0" encoding="utf-8"?> <tokenset> <language lang="en"> <token tid="ldefault">default Layout</token> <token tid="fdate">call Start Range</token> <token tid="fparty">parties</token> <token tid="fagent">agent</token> <token tid="fduration">length</token> <token tid="fdnis">service</token> <token tid="fcallid">universal Call ID</token> <token tid="fcallset">call Set</token> <token tid="cdate">call Start</token> <token tid="cduration">len</token> <token tid="cagent">agent</token> <token tid="cparty">parties</token> <token tid="cdnis">service</token> <token tid="ccallid">univ. Call ID</token> </language> + <language lang="es"> + <language lang="it"> + <language lang="ja"> + <language lang="zh"> + <language lang="fr"> </tokenset> It has an outer tag <tokenset> with inner tags <language>, which in turn have inner tags <token>. Note: Note: The naming convention adopted is as follows: the display tokens for layouts begin "l", for filters "f" and for columns "c". This is only a convention. When the application wants to display the name of a layout, it looks up its display token in the internationalization maps by name, according to the user's language preferences set in the browser. For example, lcscm.xml has the display token ldefault, the string for which (in English) is "Default Layout". To change/add strings you can supply a file called icustom.xml.this file must contain a tokenset, as described above. It is merged with the tokenset in idefault.xml. Because the file icustom.xml is read after idefault.xml, any duplicates will use your values rather than supplied defaults. For example, if your French users dislike the default translation of call ID (ID d'appel) and would prefer "ID", then supply an icustom.xml with the following: Issue 5 May,

190 Advanced Configuration <?xml version="1.0" encoding="utf-8"?> <tokenset> <language lang="fr"> <token tid="ccallid">id</token> </language> </tokenset> This text provides the override of the original value, and the column displaying call id is now labelled ID. If you are designing additional layouts, you may use any or all of the supplied token ids and these will be translated from the default file. If you wish to label things differently, you may think of additional token ids, use those in your layouts and provide translations for them in your icustom.xml. You only need to provide translations for your own languages. However, if you wish to override the default layout, you may create an lsomething.xml file with a valid layout in it, and give that layout the name "cscm". This file replaces the default layout. Note: Note: Do not call your file lcscm.xml. If you do, your file will be overwritten during upgrades. Examples To change the labels shown Create a icustom.xml file and override the ones you do not like. To change the order of the columns Copy lcscm.xml to lcust.xml. Edit the latter and delete columns or change their order. As you have not changed the name of the layout, it replaces the default one and you still have only one layout. To create additional layouts Copy lcscm.xml to lsomething.xml. Edit the latter and change the <layout> tag to give it a new name and a new display name. Make other changes as necessary. The new layout appears in alphabetical order by internal name. Create an icustom.xml. Add a translation for the display name token that you gave your new layout. 190 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

191 Usage Report Usage Report The recorder is occasionally sold to service providers with a "per use" license. Charges are levied based on the actual usage made of a recorder each month. This report can be enabled in such cases to provide the necessary billing information. Enabling the Report To enable this feature, you must add the following line to the properties file: usage.reporting=true Content The Usage Report page shows a summary of the recording modes that have been used over a specific period. The default reporting period is the previous calendar month. To generate a report for a different period, enter the date range in the calendar controls, and click Refresh. Usage of most modes is reported straightforwardly but the Bulk recording row is a count of those stations that have been Configured for Station Bulk Recording and/or Recorded using a Verint Quality Recording port at any time during the reporting period The number in the Usage column shows: the maximum size of the pool for On Demand, Meeting, Telephone Replay, Live Monitoring, Conferenced, and Unify/External using single-step conference. the total number of different stations that have been recorded in Station Bulk, Station Executive, Unify/External modes during the reporting period. The Stations Recorded column shows the individual stations and/or station ranges that were targeted. Configuration records are retained for 13 months. Each night after that period has elapsed, a background job deletes any records older than 13 months. Issue 5 May,

192 Advanced Configuration Accessing through URL: This report can be accessed directly, without having to use the normal system administration pages. To access the report by way of its URL, enter the following line in the navigation bar of your browser: Where t1 is the start time in UTC seconds and t2 is the end time in UTC seconds. Note: Note: These times are in seconds not milliseconds and do not attempt to correct for leap-seconds. Unless an administrator made configuration changes within a few seconds of midnight, this will not affect reporting on monthly boundaries. The names of recording modes and other information that can be localized are returned according to the language preferences established for the interface. If localized terms are not available, the returned values are in English. Accessing the Usage report in a log file If you request the usage report by way of its URL and the request is successful, the recorder writes the usage data to a log file called usage.log in the /opt/witness/logs directory beneath the install path. You can access and view this log file in any text viewer. Its content is the same as the report returned for the URL request. Each time you request a usage report, a new log is created that overwrites the previous one. If a request for the Usage report is not successful, no log file is written for it. You should examine the return value of the URL request for an indication of the error conditions (bad time parameters, configuration has been tampered with). 192 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

193 Selective Record Barring Selective Record Barring Recent legislation changes mean that some customers may need to block recording of calls to or from specific area codes (e.g. California). Configuration It is possible to bar recording of calls to or from certain numbers. To configure such a recording bar, add the property file entry: recording.barred=<regular expression> Where <regular expression> is a regular expression that will match the digit strings to be barred from recording. Tip: Tip: See for instructions on how to form a regular expression. Example The following example shows how to bar calls to or from area codes 234, 567 and 890 where the recorder is situated in area code 234 (which therefore does not have a 1 in front of it, unlike the others which may or may not). The trailing periods (there are seven of them) are important as this forces the pattern to match only numbers with 7 digits following the area code. recording.barred=((234) (1?567) 1?890))... Any recording that is barred due to matching the digit pattern specified will cause an INFO level message to appear in the log file. Limitations 1. This feature applies to Station Bulk, Station Executive, Unify/External Control (Station Bulk) and Conferenced modes only (On Demand, Meeting and Externally controlled Conference mode are not affected). 2. To bar incoming only or outgoing only, first determine the digit patterns that are used. You may be able to change the outbound dial plan to always prefix with a 1. The "1?" Issue 5 May,

194 Advanced Configuration matches calls with or without a preceding 1. Remove the? to require a 1 before the pattern is matched. 3. Behavior differs slightly between the Conferenced and Service Observe modes. The former break each recording into separate segments whenever the parties on a call change e.g. consultation call becomes 3-way conference call. The record/bar decision can therefore be applied on each segment and an internal consult call will therefore be recorded even though the resultant conference call (involving a barred party) will not. Service Observe, however, does not break the media stream at this point and hence a single recording segment containing the consult call and the three-way call is the norm. In this case, if a barred party is present at all (i.e. would appear in the list of parties tagged on the recording) then the entire segment in this case including the consult call is deleted rather than stored. 4. Note that the Service Observe modes may record calls (as Station Executive mode does for all calls) that are in progress but then delete the recording and associated data file at the end of the call. 194 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

195 Appendix A: Technical Reference This appendix provides technical details about the Verint ContactStore for Communication Manager system. The main sections in this appendix are: Recording files on page 196 Internal Database on page 197 Recorder Interfaces on page 199 Issue 5 May,

196 Technical Reference Recording files Call segments are stored in an industry standard wav file. When each call is completed and as each recorded call segment becomes available, the recorder updates its local database with a record of the call segment. These files are stored in a hierarchy of folders beneath /calls. Every recording results in one or more: wav file xml file WAV files The wav files contain the actual audio of the recording. You can double-click some wav files to play them directly. Others are in audio formats that are not directly supported by Microsoft's Media Player. This applies to most recordings made by this recorder. These must be converted into a supported format before they can be played. Since the recorder's Search and Replay application does this conversion automatically, you do not need to access these files directly. XML files The xml files contain details about the recorded call segments. Although most users typically search against the recorder's database of calls, you can view these files directly in a browser if required. Within each xml file there is: All the details known about this recording. Most of the information, but not all, is inserted into the calls database. Some of the information is only of interest for diagnostic and maintenance purposes. Start and end time in ISO format giving local time and offset from GMT. 196 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

197 Internal Database Internal Database If you have retained all of the xml and wav files as described above, then you have kept all of the details about the recordings you have made. However, the system uses an industry standard database (PostgreSQL) to hold this information in more readily accessible forms. This database is located on the recorder itself. The database stores details of the recordings as well as details of the recorder's configuration. Recording details The call details database uses approximately 1KB per call (in the absence of user defined fields). To allow you to search for calls easily, the details of recordings (except those made for and copied off to a Quality monitoring server) are inserted into this database. It contains one record for each call segment recorded and additional records for each party on the call and each owner of the call. The information stored for each call is: A unique reference for the recording The start date and time The duration of the recording The recording mode that was used to make the recording The Communication Manager's Universal Call Identifier (UCID) The name and number of the parties on the call-where this was available to the Communication Manager (through ANI or CLI) at the time of the call The agent number and name of an agent involved in the recording if one was logged on at the station being recorded The direction of the call (incoming or outgoing) The owner(s) of the recording The service or vector directory number (VDN) name where available Additional fields provided by external controllers The DNIS if present (Conferenced mode only). Stored as user defined field "spare2" Issue 5 May,

198 Technical Reference Configuration details Several tables hold details of system configuration, such as port assignments, file paths, timeouts and user authorization rights. 198 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

199 Recorder Interfaces Recorder Interfaces The interfaces supported by the recorder are described below (working clockwise round the diagram, starting at top left). HTTP Interfaces Offered The recorder uses the tomcat (see web servlet container to offer a number of services via HTTP and/or HTTPS (on ports 8080 and 8443 respectively). These provide Administration Interface This provides administrators with access to configuration and status monitoring pages. Issue 5 May,

200 Technical Reference Search Interface End users access this to search for call recordings that match specific criteria. Replay (Retriever) Interface End users and other applications (such as Viewer, Vision and Archive) use this interface to retrieve the voice content of a specific recording. AudioServer Interface If at least one port on the recorder is licensed for and assigned to the Telephony Replay pool, then the recorder supports the AudioServer Interface. The login page for users entitled to replay allows entry of a telephone number on which they can be reached. If they do not enter a number, the browser replays through the soundcard. You can also use separate Replay applications. Applications that can use this interface include: Viewer and eware Vision Call Details Interface The details about a recording can be provided on request. This is used by the bulk export feature of the integral search and replay application and the Central Replay Server to populate its database. Communication Manager DMCC TSAPI (Avaya CT) The recorder interfaces to the Avaya components via several mechanisms: DMCC runs on an AE Server and provides softphone registration and signalling services. Internally, the Verint software uses JTAPI but this is layered on top of TSAPI and hence the recorder server exchanges TSAPI messages with an AE Server if Conferenced recording is used or if an external controller asks the recorder to establish single-step conferences. 200 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

201 Recorder Interfaces Audio over RTP The softphones on the recorder terminate RTP streams over which the audio to be recorded flows. These sockets are connected to whichever VoIP resource (for example, Media Processor) is providing the conferencing bridge through which the call is recorded. The recorder uses ports upwards, allocating two ports per recording channel. equality Balance V5 or V7 in Selective Recording Mode Recording Command Interface The recorder can establish a TCP/IP connection to port 1415 on one or more equality Balance servers. The Quality monitoring application instructs the recorder to record specific calls; to replay calls via the phone and live monitor calls. Recordings Written to Disk As the recorder completes each voice recording that an equality Balance server requests, it copies the audio to a fileshare on the Quality Monitoring server. This uses a fileshare as specified in the quality monitoring server's.ini file. Username and password for the fileshare are configured on the recorder. Recordings Read from Disk As equality Balance instructs the recorder to replay calls, the recorder reads the audio from files it previously placed on the Quality monitoring server's disk. This uses the same fileshare or local path (if co-hosted) as above. equality Balance V7 in "ContactStore Plus" mode The recorder can interact with equality Balance V7 in one of two ways - either in the same "selective recording" mode as described above OR in what is known as "ContactStore Plus" mode - using the eqconnect Interface below to advise the quality application of all the bulk recordings it has made eqconnect Interface The recorder can use an HTTP interface to communicate with eqconnect if this is installed on an equality Balance (V7) server. (This defaults to port 3020). This is used to advise the Balance server of agent logon/logoff and recording details. Issue 5 May,

202 Technical Reference Replay Interface equality Balance uses Viewer to replay calls made on this recorder. Other Recorders Recorders establish links with each other when acting in Master/Slave and Primary/Standby topologies. These are TCP/IP socket interfaces. Master/Slave Recorder Control A recorder configured for Conferenced recording mode can control additional 'slave' recorders. Each slave establishes contact with TCP/IP port 1416 on the Master. Primary/Standby Recorder Interface Standby recorders establish contact with TCP/IP port 1414 on the primary recorder. The primary provides configuration data to the standby and instructs it to take over should it sense a fatal error. Two such links must be established over separate paths. The recorders exchange polling messages once a second. External Control Interface The recorder supports a simple TCP/IP command interface. This allows other applications to control recording directly and/or add further user defined data fields to recordings. This protocol supports START, STOP, TAG and BREAK (seamless stop and start) commands. Database Upload Interface When configured with either a Central Replay Server or a Viewer database, the recorder uploads details of calls into database on the apropriate server. To Central Replay Server To upload call details to the Central Replay Server (CRS) a recorder uses a web service on the CRS, to inform it that a new recording is available. The CRS in turn uses a web service on the recorder to retrieve the details. Both of these interactions use port Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

203 Recorder Interfaces To Viewer database When deploying Viewer, the recorder uploads call details to a central Microsoft SQL Server database over this link. The recorder acts as a standard Microsoft SQL Server client application. It uses the TDS protocol, which uses TCP/IP port 1433 by default. This can be overridden with a properties file setting. Summary Interface Protocol Local Remote Direction HTTP - Admin and Replay TCP 8080 Inbound HTTPS - Admin and Replay TCP 8443 Inbound DMCC TCP 4722 Outbound TSAPI TCP 450, 1050 Outbound RTP UDP Bidirectional Selective Quality Control TCP 1415 Outbound Selective Quality Fileshare NTLM Outbound ContactStore Plus Quality TCP 3020 Outbound Inter-Recorder communication TCP Towards master Central Replay Server TCP 8080 Bidirectional Viewer Database TCP 1433 Outbound Issue 5 May,

204 Technical Reference 204 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

205 Appendix B: Troubleshooting This appendix covers two areas: general troubleshooting tips and some specific common issues: The main sections in this appendix are: Hints and Tips on page 206 Specific Problems on page 208 Issue 5 May,

206 Troubleshooting Hints and Tips Where to Look for Clues When problems occur, check the following: ed Alarms and Events. If you have been using the settings to have alarms and events forwarded to one or more addresses, you should check these carefully. As well as checking the contents of messages that you have received, also check for days when the nightly log file purge message has not been received. Alarms Page. This page within the administration application provides a wealth of information on problems that the system has detected. Review the alarms carefully. If the problem is not immediately apparent, consider viewing all alarms, including those that have previously been cleared. It may be that someone has cleared an alarm without addressing it or realizing its significance. Log Files. Check for errors being reported in log files within the following directories beneath your installation directory: /opt/witness/logs /opt/witness/tomcat5525/logs Determining Current Version When reporting problems you should state precisely which version of software you are running. To determine this, click on the System Settings > License page. Note the precise version number shown. Application Logs The Verint ContactStore for Communication Manager writes log files to /opt/witness/logs. The current day's log file is called cscm.log. At midnight the current log file is closed, renamed to cscm.log.<date>, and a new log file opened. These log files are automatically purged after 30 days by default but this can be overriden with a properties file entry. 206 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

207 Hints and Tips Setting the server log level You can set the level of messages logged by the Verint ContactStore for Communication Manager to DEBUG, INFO (the default), WARN, ERROR or FATAL in one of two ways. Permanently from next restart To change the level permanently, enter the following line in the properties file cscm.properties and restart the Verint ContactStore for Communication Manager service to have it take effect. log.level=debug If you change the log level in the properties file, it remains set on subsequent restarts. When logging at DEBUG level, note that the log files grow very quickly and can overflow the disk if left at this level. Temporarily, immediately To change the level temporarily, without restarting the service, simply use a browser to request the URL: using the name of your server. Alternatively, if you do not have access to the web administration screens, but do have access via secure shell, execute the following command: perl /opt/witness/bin/loglevel.pl DEBUG You do not have to stop recording in order to change the logging level. To set it back again, enter the same URL, replacing DEBUG with INFO. The command is case-sensitive. Using this method changes the log level temporarily. It will revert to normal the next time that the system is rebooted. Tomcat Logs Verint ContactStore for Communication Manager uses the Tomcat web servlet container, which writes log files to /opt/witness/tomcat5525/logs. Remote Access You are strongly advised to provide remote access to the server to aid fault-finding. Issue 5 May,

208 Troubleshooting Specific Problems System Administration page problems You may encounter problems as you access and use the System Administration application. This section lists those problems and suggests steps to take to correct them. Cannot access the System Administration pages Cannot log in If you cannot access the System Administration pages, try the following: Ping the server to confirm that connectivity is possible. If not, trace the network connections between client and server and double-check the server's IP address, default gateway etc. Use the numeric dot notation IP address instead of the hostname. If this works, then the hostname is wrong or cannot be translated by your DNS services. You may need to use a fully qualified node name, such as recorder.bigco.com. Use the browser installed on the server itself to access the application at If this works, then the problem is in the network between server and client. If it does not work, then the problem may be with the Tomcat web server. If you have trouble logging in, double-check the state of Caps Lock and ensure the password is being entered with the correct case. If you can log in under another account, reset the password of the account having problems. If nothing happens when you click the OK button, check that your Internet Explorer settings allow javascript to run. See ActiveX Control Download on page 138. These symptoms have also been seen when trying to access a server with an underbar in its node name. Note that this is not a valid IP name and should be changed. Web pages appear "messy" If the forms on the System Administration pages and/or Search and Replay page appear to be out of alignment, or fields are truncated, check your desktop properties. If set to Large Fonts (120dpi), this will impact all website accesses not just the Recorder pages. 208 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

209 Specific Problems To correct display issues: 1. Right-click on your desktop and select Properties. 2. On the Settings tab, click Advanced and set the DPI Setting to 96dpi. On new laptops, especially those with 1400x1050 resolution screens, the DPI is often set to 120dpi. Using too small a window or too large a type (View > Text Size in Internet Explorer) can make the columns and headings on the web pages wrap where they would not normally. Connectivity alarm problems Invalid entries in any one of the parameters used to define the settings will result in errors. To check this: Try the settings you are using in a standard mail client, such as Outlook. Send a message using the account specified to prove that the settings are valid. If messages have been working and then stop without any of the settings changing, verify that nothing has changed on the mail server. This problem occurs, for example, if your password has been reset or changed on the mail server. If the recorder is not sending messages, it may be because it is not able to access the SMTP server. Check the network connections to the recorder. Recorder to AE Server Problems The recorder communicates constantly with the AE server. If this link fails, alarms are raised. If you suddenly get multiple alarms, typically those reporting problems with softphones timing out registration, this link might have failed. Ping the AE Server to check the link. Recorder to Quality Monitoring server problems If the problems reported include mention of the file-share: 1. Verify that the Quality Monitoring server is running. 2. Verify that the username and password previously in use are still valid. Try to access the file-share from another machine using the username and password provided to the recorder. Issue 5 May,

210 Troubleshooting Recorder to Viewer server problems If you have problems connecting to the Viewer server: 1. Verify that the Viewer server is running. 2. Verify that the SQL Server is active on it. Search and Replay problems For most problems with Search and Replay, consider the following diagnostic approaches to narrow down the cause of the problem: 1. Search for a different call, for example, one that is more recent or older, shorter or longer. 2. Log in as a different user with different replay restrictions Cannot access the replay application If you cannot get to the login page, try accessing the page from a different machine: 1. From the same side of any firewalls. 2. On the same LAN if you are having problems with WAN access 3. From the same sub-net, if having problems from a different sub-net 4. From the recorder itself, if having problems from the same sub-net. User replay restrictions do not work If you have given a user account replay rights over a number of addresses but the calls from these stations are not listed when you enter a valid search that should include them, check the recording ownership. When an agent has logged on to a station that is being recorded, the calls recorded are "owned" by the agent number not the underlying station. Give the user replay rights to the range of agent IDs who have been using the stations in question. In fact, in most cases, you should restrict access to a set of Agent IDs rather than station numbers. Problems downloading ActiveX control If you see error messages relating to ActiveX controls being downloaded-or blocked from being downloaded, your security settings may be too restrictive. See ActiveX Control Download on page 138 for more information. 210 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

211 Specific Problems Problems displaying ActiveX control Cannot log in If the ActiveX control downloads but displays a red cross on white background at top of page, upgrade to Internet Explorer 6.0. These symptoms have been seen on Internet Explorer 5.0. If you see the login page but cannot get past it: 1. Verify that Caps Lock is off and that you are entering the password with the correct case. 2. Log in as a different user 3. Confirm the spelling of your log in name with the system administrator and check that your account is still configured in the administration pages. 4. Ask the system administrator to reset your password. Log in with a blank password and change your password when redirected to the Change Password page. Search returns no calls If you get to the search page but no calls are returned when you perform a search: 1. Broaden your search criteria to confirm that you can at least find some calls. Start by requesting calls from any parties for today. If that shows no calls, extend the time period. 2. Try setting the date range back to at least the time you know you have seen call records for in the past. 3. Check that the system administrator has given you access to the correct calls. Your search and replay restriction may be wrong or too narrow for the search you are attempting. 4. Confirm that calls are being recorded. Follow the troubleshooting guidelines for recording problems if you suspect that the system is not actually recording or processing any calls. Calls listed but cannot play them If you can see the list of calls that matched your search criteria, but cannot actually play them, look at the area at the top of the browser page where the "graph" of the audio normally shows and match your symptoms to one of the following: Issue 5 May,

212 Troubleshooting No Audio "graph" This means that the call has not been retrieved from the recorder or DVD disk or has not reached the client PC. 1. Check the server logs for errors. 2. Note the call's 15 digit reference number (shown if you hold the mouse pointer just to the right of the radio button that you click to retrieve the recording. Search for that wav file in the calls path to confirm that the recorded file exists. 3. Check connectivity and available bandwidth to the client PC. Audio graph stops in mid call This implies that the transfer of data from the Recorder to your client PC has been stopped or interrupted. 1. Request the same call again. There may have been a temporary network problem. 2. Request a different call. If the problem is only with one call, you may have a corrupt file on your hard disk. 3. Request the problem call from another PC on the same network. If the other PC can retrieve it successfully, assess the differences between the two client PCs; the problem is most likely at the client end. 4. Request the problem call from different sub-nets, ideally working closer to the recorder. 5. Request the call from the recorder server's own browser. If this works and the others don't, then the problem is likely to be in the network between server and clients. Audio graph appears but no sound The audio file has reached the client PC successfully; the problem is most likely to be with the PC's multimedia setup or current settings. 1. Verify that the PC has a sound card. 2. Play a wav file through Media Player or similar application to verify that that the sound card is set up correctly. 3. Adjust any hardware volume and/or mute controls on the speakers/headphones. 4. Double-click on the icon in the system tray at the bottom right-hand corner of the screen to verify that the PC's software volume controls are not set to mute or very low. 5. Ensure you are not running any other programs that may be locking the sound card exclusively. If in doubt, shut down all other programs. 6. Try another similar PC. If that works, look for differences in the multimedia setup of the two PCs. 212 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

213 Specific Problems No New Recordings Playable If you can replay old recordings but not newly made calls, there may be a problem with the recording and/or storage components of the system. Follow these steps: 1. On the Status > System Overview page of the Recorder's System Administration application, look at the counts for total calls recorded and calls recorded today. 2. Use Bulk recording mode to make a test recording. 3. Complete the recording and hang up. 4. Return to the Status > System Overview page and note the Total call segments recorded today and Total call segments recorded to date (or since restart, if today). These counts should have increased by at least one, the recording that you just made. If the counts have increased, the recorderingis processing recordings. This is probably a search/replay problem. See earlier sections for help. Note: Note: If these counts have not increased, the recording has not been successfully compressed and stored on the Recorder or inserted into the call details database. Do the following: 5. Look for alarm messages that indicate problems with the recording channel or with file read/write or rename functions. The error message should indicate whether disk space or a directory access problem is the cause. Check that.wav files are appearing in the latest folder beneath the calls path as recordings are made. 6. Check disk space in all partitions. If any of these is 0 or less than 50MB, this may be the problem. Check for build up of log files. Check that the call details database hasn't exceeded the available space. Consider reducing the number of months of calls kept-use the purge settings on the Server configuration page to adjust this. 7. Look for alarm messages that indicate licensing problems. The recorder will not process any new calls if you have changed the MAC address, tampered with license settings or are running on a time-expired license. In all cases, you should obtain a new license key. 8. Check that the Tomcat server is running (cscm service). 9. Check for messages in the log files. 10. Reboot the server and watch for error messages on startup. Poor Audio Quality on Telephone Replay If the network configuration is correct and there are no problems with its function, the most likely cause of this problem is that the recorder or the network is overloaded. To look for a problem on a managed network switch, you should look at the diagnostics and configuration details. You should see ZERO errors on all ports. Any port showing more than 1 packet error in 10,000 is suspect and must be looked at. Issue 5 May,

214 Troubleshooting If the C-LAN, MedPro and/or recorder ports are failing to auto-sense full/half-duplex properly, you can force each port to either full or half duplex so as to reduce the error count to zero. Note: Note: Even though a port may show an error rate of less than 1 in 100 packets, the error counts are deceptive. A single packet error can trigger a full/half duplex negotiation during which all packets are lost in the servers, but none of these show as errors on the switch. If your error counts are zero on all ports, then we must also consider overload of the recorder as a possible cause. You should monitor the CPU load of the recorder during busy hours. Replay and live monitor are very sensitive to overload. Recording may be unaffected but if the CPU load is too high, audio quality on replay can suffer. Similar problems have also been seen on multi-cpu AMD Opteron servers. This is caused by an unstable system clock, which is addressed in RedHat Versions 4 and 5. Recording Problems Partial recording problems Since no hardware component in the system is dedicated to specific ports, any hardware problem is likely to affect all recordings equally. Therefore, if some calls are being recorded and are playable but others are not, the problem is probably in the configuration. 1. Recording Mode versus Recording Channel? Determine whether your problems relate to all channels of one or more recording modes or just to certain ports. 2. Check the configuration pages for the affected recording mode(s). 3. Calculate the range of stations carefully. For example, to is a range of 11 addresses, not Use the Status > Port Status page to observe the ports on the recorder during your test calls. The ports go should go from idle to active and back again. Quality Recording Check that the Quality recording ports are set to use the same codec as any bulk recording you are doing on the same phones. 214 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

215 Specific Problems Meeting Recording Cannot Enter Owners The recorder requires rtp-payload signalling as described in Configuring tone detection on page 75 in order to interpret dialled digits. Check that both IP phones and Digital phones are configured for this signalling mode as IP phones will default to it whereas digital phones may not. Station Bulk Recording Cannot Enter Delete Command The recorder requires rtp-payload signalling as described in Configuring tone detection on page 75 in order to interpret dialled digits. Check that both IP phones and Digital phones are configured for this signalling mode as IP phones will default to it whereas digital phones may not. Station Executive Recording Cannot retain recordings The recorder requires rtp-payload signalling as described in Configuring tone detection on page 75 in order to interpret dialled digits. Check that both IP phones and Digital phones are configured for this signalling mode as IP phones will default to it whereas digital phones may not. Live Monitor Cannot monitor 1. Check that you have created the appropriate numerical user account so that the station/phone dialling in is recognized as an authorized user. 2. Check that the live monitor ports are set to use the same codec as the ports you are trying to monitor. 3. The recorder requires rtp-payload signalling as described in Configuring tone detection on page 75 in order to interpret dialled digits. Check that both IP phones and Digital phones are configured for this signalling mode as IP phones will default to it whereas digital phones may not. Poor Audio Quality See Poor Audio Quality on Telephone Replay on page 213. The causes are the same. Issue 5 May,

216 Troubleshooting 216 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

217 Appendix C: Alarms This appendix provides details of the alarms that can be raised by the system.. The main sections in this appendix are: Alarms on page 218 Alarms Table on page 219 Issue 5 May,

218 Alarms Alarms The recorder may generate the following Alarm or Event notification messages. These events are: shown on the Status > Alarms/Events page sent in messages as specified on the System Settings > Server page reported via SNMP logged to the recorder's log file cscm.log 218 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

219 Alarms Table Alarms Table Within messages, the strings XXX and YYY represent a specific parameter such as a station number, an IP address etc. The table shows both the English text of the message (that appears on the Alarms page) and the underlying resource string that appears in the log file. Entries within the table are sorted according to the text of the message. Severity Log File Entry English Alarm text Comment What to Do Warning alarms.binaryuui One or more User to User Information fields are in binary format. This is not supported. See call XXX for example. Major alarms.callspath.invalid Call storage path is invalid. Please change it under Settings > Server. Major alarms.cmapi.down Device, Media AND Call Control API not running on YYY. Reason: XXX Major alarms.cmapi.up Device, Media AND Call Control API running on XXX. Major alarms.dapi.down Communication Manager Data Link DOWN. Please check the current status on System Overview page. Reason: XXX Major alarms.dapi.up Communication Manager Data link UP. Major alarms.dapi.aesvcsconfig Cannot use DAPI. Please check you have added XXX to the H.323 Gatekeeper list on the web administration Switch Connections screen. Only text fields are supported. Separate partition preferred. Major failure of DMCC services. DMCC services are restored. The link remains down unless advised otherwise. Alarm will repeat daily until resolved. DAPI services are restored. See AE Server manuals Change to text fields or stop attempting to store UUI Set this to a valid path. Check connectivity to AE Server. No action required. Check connectivity to AE server. No action required. Configure the CLANs as H.323 Gatekeepers. Major alarms.database.cannotinser t Cannot insert details of recording XXX into database. Check disk space on the partition holding the call details database. Major alarms.disk.full Disk full on partition 'XXX 'Check log files. If calls partition, check for files that cannot be purged. Delete some files to make space. Issue 5 May,

220 Alarms Severity Log File Entry English Alarm text Comment What to Do Warning alarms.disknearlyfull Disk nearly full on partition 'XXX'. Only YYY MB free. Major alarms.eware.diskload Error advising eware of disk insertion. Reason: XXX Major alarms.eware.linkdown eware Server Link DOWN. Reason: XXX Check log files. If calls partition, check for files that cannot be purged. Delete some files to make space. Check connectivity to SQL server. Check connectivity to SQL server. Major alarms.eware.linkup eware Server Link UP. No action required Major alarms.eware.tar Error adding archive details to eware. Reason: XXX Check connectivity to SQL server. Major alarms.file.copyfail Failed to copy file XXX. Check disk space, path, security. Major alarms.file.copyretry Retrying file copy of XXX. Minor alarms.file.deletefailed Failed to delete file XXX. Reason: YYY. Major alarms.file.iocopyfail Failed to copy file XXX. I/O Error: YYY. Minor alarms.file.wavwritefail Failed to write file XXX.wav. I/O Error: YYY. Minor alarms.file.compressfailed Failed to compress audio file for recording XXX. Reason YYY. Major alarms.file.xmlwritefail Failed to write XML file XXX. Reason: YYY. Warning alarms.file.notindb XXX recordings up to and including YYY were not stored to database on last shutdown. Warning alarms.jtapi.backingoff Backing off Single-step Conference control as other recorder is establishing calls. Major alarms.jtapi.backlogcleared Single-step Conference Queue Backlogged. Flushing. Major alarms.jtapi.backlogged Single-step Conference Backlog Cleared. Implies last shutdown was not "clean". If no standby recorder is present, can imply spurious call details being advised by Avaya CT. Queue of requests is too long. Queue of requests now acceptable length. No action required Depends on reason Depends on reason Depends on reason Depends on reason Depends on reason Shut down service rather than kill. Check if standby did take over. If not, report problem. Check system loading and increase capacity if needed. No action required. 220 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

221 Alarms Table Severity Log File Entry English Alarm text Comment What to Do Warning alarms.jtapi.conffailed Single-step Conference Failed: XXX Very occasional failures are nothing to be concerned about. Regular failures should be reported. Increase number of retries in properties file. Check for latest versions of Avaya CT\TSAPI. Major alarms.jtapi.configchanged Avaya CT Configuration changed. Dropping previous connection. Minor alarms.jtapi.csta44 Single-step Conference failed on address XXX. CSTA Error 44. Check that no other AE Server is controlling this call. Major alarms.jtapi.error Avaya CT Services on XXX reporting error: YYY Major alarms.jtapi.heartbeatfailed Avaya CT Services failed. Heartbeat lost. Major alarms.jtapi.invalidparams Invalid Service, username or password parameters for Avaya CT: XXX Major alarms.jtapi.invalidskill Invalid Skill Group: XXX User changed configuration details, Always means that another application is trying to control the call. System should attempt to restart Avaya CT services automatically.. No action required. Stop using Follow the call options or stop the other application. Depends on error shown. Report if recurring, Check all parameters. Check that this is a valid skill hunt group. Major alarms.jtapi.invalidvdn Invalid VDN: XXX Check that this is a valid VDN. Major alarms.jtapi.notaskill Not a valid skill group (or Avaya CT cannot reach switch) Check that this is a valid skill hunt group. Major alarms.jtapi.observationende d Observation ended Report problem. Major alarms.jtapi.observer Avaya CT Observation of XXX reports error:yyy Report problem. Major alarms.jtapi.outofservice Out of Service Check Avaya CT services. Major alarms.jtapi.overflow Conferenced Recording Mode failed to record a call because all ports were busy. Increase number of concurrent recordings allowed on Conferenced mode. Issue 5 May,

222 Alarms Severity Log File Entry English Alarm text Comment What to Do Major alarms.jtapi.portsavailable Conferenced Recording Mode now has XXX ports available. Warning alarms.jtapi.portslow Conferenced Recording Mode has less than XXX ports available. Major alarms.jtapi.providerok Avaya CT Services on XXX UP No action required. Increase number of concurrent recordings allowed on Conferenced mode. No action required. Major alarms.jtapi.shutdown Service shut down Check Avaya CT services. Warning alarms.jtapi.takingover Taking over Single-step Conference control as other recorder failed to establish call. Info alarms.logpurge Purged XXXKB of old log files. Now YYYMB free. Minor alarms.monitor.codecwrong Live Monitor failed due to incompatible audio formats. Monitor ports and monitored ports must have identical Audio Format settings Major alarms.master.linkerr Error establishing link to Slave recorder on XXX: "YYY Major alarms.master.serversocket Error creating socket for slave recorder on port XXX. Reason: YYY. Major alarms.master.timeout Slave recorder timed out: XXX at address YYY Major alarms.primary.linkerr Error establishing link to Standby recorder on XXX: "YYY Major alarms.primary.serversocket Error creating socket for standby recorder on port XXX. Reason: YYY. Major alarms.primary.timeout Standby recorder timed out: XXX at address YYY See alarms.jtapi.backing off Nightly message. Check standby recorder. No action required if received. If not received, check recorder is running. Reconfigure monitor ports to use same codec as other modes. "Depends on reason shown. Depends on reason shown. Check slave recorder and network paths to it. "Depends on reason shown. Depends on reason shown. Check standby recorder and network paths to it. 222 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

223 Alarms Table Severity Log File Entry English Alarm text Comment What to Do Major alarms.properties.notanumbe r Property XXX in properties file is non numeric (YYY). Using default value instead. Correct the entry in the properties file. Major alarms.queue.acceptable XXX Job Queue backlog reduced to acceptable level. Currently YYYms. Major alarms.queue.copy Failed to copy to XXX. Reason:YYY. NOTE: Further errors with the same root cause on the same path will only show here once every 24 hours. Check the log file if in doubt. Major alarms.queue.slowjob XXX Job Queue individual slow job - took YYYms. Major alarms.queue.tooslow XXX Job Queue backlogged. Delay currently YYYms but may go higher. Minor alarms.rtp.packetloss Unacceptable packet loss from XXX. Major alarms.rtp.nopackets No audio packets received on call from XXX. Call failed or no network path to recorder. Major alarms.softphone.beepstart Port XXX. Failed to start beeptone. Reason: YYY Major alarms.softphone.beepstop Port XXX. Failed to stop beeptone. Reason: YYY Will cache files until problem is resolved then copy all outstanding. Occasional message not a problem. May indicate overload. Reported once per day per IP address but could be occurring on all recordings from that address. See Avaya support site re PSN #345U - TN570C and TN570D Expansion Interface boards log chronic fiber alarms Internal error Internal error No action required Determine why recorder cannot write to the share. Report frequent or very long delays. Consider faster hardware or reduce load. Check network path to recorder. If several different addresses report problems at the same time, identify common path. If many addresses report at same time, could be CPU overload at recorder. Check TN570 boards are correct vintage. Report problem. Report problem. Major alarms.softphone.buttonlook up Port XXX. Error looking up button Id. Reason: YYY. Softphone misconfigured. Check softphone configuration is as per installation section. Issue 5 May,

224 Alarms Severity Log File Entry English Alarm text Comment What to Do Major alarms.softphone.buttonmissi ng Port XXX. Unusable as button YYY is missing from softphone. Softphone misconfigured Check softphone configuration is as per installation section. Major alarms.softphone.callstopped Port XXX. Call dropped as it exceeded maximum permitted duration. Major alarms.softphone.cmapidown Port XXX. DMCC services not available. Recording port has been active for several hours. Consider setting recording mode to release call on dropping to one other party. Verify connectivity to AE services and that DMCC is running. Major alarms.softphone.endrecfaile d Port XXX. Error ending recording. Reason: YYY. Internal error Report Problem Major alarms.softphone.hookswitch Port XXX. Error setting hook switch. Reason: YYY. Internal error Report Problem Major alarms.softphone.inservice Port XXX restored. No action required Minor alarms.softphone.invalidtarge t Port XXX - Invalid target for recording: YYY Quality app asked to record invalid station. Check Quality server configuration. Major alarms.softphone.mediacontr ollistener Port XXX. Cannot add/remove MediaControlListener. Reason YYY. Internal error Report Problem Major alarms.softphone.notarget Port XXX. Service observe failed as no target station set. Major alarms.softphone.nullpointer Port XXX. DMCC event 'YYY' fired with null pointer. Major alarms.softphone.ownerreg Port XXX. Error unregistering owner. Reason: YYY. Internal DMCC error related to a particular recorder port. Internal error Check configuration Report occurrences with a copy of your log files. Report Problem. Minor alarms.softphone.outofservic e Port XXX out of service. Should recover within a few seconds. Depends on reason shown before this alarm. Major alarms.softphone.play Port XXX. Error playing file. Reason: YYY. Depends on reason shown. Minor alarms.softphone.processfail ed Error processing file XXX. Reason YYY. Depends on reason shown. Major alarms.softphone.recordstartf ailed Port XXX. Failed to enable recording. Reason: YYY Depends on reason shown. 224 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

225 Alarms Table Severity Log File Entry English Alarm text Comment What to Do Major alarms.softphone.recordtime out Port XXX. Recording failed to start. Resetting port. Depends on reason shown. Major alarms.softphone.registration failed Port XXX Registration failed. Reason: YYY Depends on reason shown. Major alarms.softphone.regtimeout Port XXX. Registration timed out. Major alarms.softphone.rtpthread Port XXX. RTP handler failed. Reason: YYY. Check connectivity to AE Server. Depends on reason shown. Major alarms.softphone.setuprecor ding Port XXX. Error setting up recording. Reason YYY. Depends on reason shown. Warning alarms.softphone.shortpacke t Port XXX. VoIP packet interval of YYYms less than recommended 60ms. Default settings of 20 or 30ms are not efficient for recording. Configure codec set and network region as per installation instructions. Warning alarms.softphone.sodropped Port XXX. Service observe dropped unexpectedly while in state 'YYY'. Major alarms.softphone.sofailed Port XXX. Service observe failed despite multiple retries. Major alarms.softphone.soachieved Port XXX. Service observe achieved on YYY. Minor alarms.softphone.sodenied Port XXX. Service observe denied trying to observe YYY. Check station is allowed to be service observed, the station is present and no other service observe active. Try manual service observe. Also check that G729A and G711MU codecs are enabled for the recorder's softphone region. Warning alarms.softphone.sotimeout Port XXX. Service observe setup timed out. Trying to observe YYY. Check for resource overload on switch. Check COR of softphone and target. Check for other observers. No action required. See text of alarm. As above. Major alarms.softphone.sscdroppe d Port XXX. Single-step conference dropped unexpectedly while in state YYY. Report problem if more than very occasional occurrences. Issue 5 May,

226 Alarms Severity Log File Entry English Alarm text Comment What to Do Major alarms.softphone.sscfailed Port XXX. Single-step conference failed despite multiple retries. Major alarms.softphone.ssctimeout Port XXX. Single-step conference setup timed out. Trying to conference in to YYY. Consider increasing number of retries. Check system loading. Major alarms.softphone.startrecfail ed Port XXX. Recording failed to start. Reason: YYY Depends on reason shown. Major alarms.softphone.stopplaying Port XXX. Error stopping file playing. Reason: YYY. Depends on reason shown. Major alarms.softphone.stoprecordf ailed Port XXX. Error disabling recording. Reason: YYY. Depends on reason shown. Major alarms.softphone.timedout Port XXX. Timed out when in state YYY. Major alarms.softphone.ttd Port XXX. Failed to set up touch tone detect. Check you have OUT_BAND signalling configured. Failure Reason: YYY. Warning alarm.softphone.brokenflutter Port XXX. Feature failed on button YYY. Report if more than very occasional occurrences. Configure signaling in accordance with installation instructions. Check COR, switch resources. Major alarms.softphone.unregistere d Port XXX. Unregistered. Reason: YYY. Can imply standby recorder has registered this phone. Check if standby is active. Major alarms.softphone.unregisterf ailed Port XXX. Error unregistering. Reason: YYY Depends on reason. Warning alarms.softphone.userreset Port XXX. User 'YYY' reset the port. Also logged to audit trail. No action required. Major alarms.softphone.voiceunitlis tener Port XXX. Cannot add/remove VoiceUnitListener. Reason: YYY. Internal error. Report problem. Major alarms.standby.invalidstns Invalid Station list "YYY" received from primary for station pool "XXX Major alarms.standby.noswitch No link to Communication Manager, All ports have failed. "Internal error Report problem. Check AE server is up. 226 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

227 Alarms Table Severity Log File Entry English Alarm text Comment What to Do Major alarms.standby.notviable Recorder fatal error. Recorder will request that standby takes over if one is present. Major alarms.standby.primarynotok Master recorder requests that Standby(s) take over. Minor alarms.unify.ioexception File I/O error updating call details of XXX. Reason: YYY. Minor alarms.unify.parserror Error parsing Unify/External control information. XXX is not a number. Minor alarms.unify.parseexception Parsing error updating call details of XXX. Reason: YYY. Major alarms.url.badport Invalid IP port number for recording mode: XXX, Set to: YYY Major alarms.url.general Error on link to 'XXX': YYY Major alarms.url.linkup Link established with XXX server on 'YYY Major alarms.url.nourl Ports are allocated to XXX recording but no url is set to communicate with the server. See next error message. Check fatal error causing switchover. Depends on reason. Correct external controller command. Depends on reason. No action required. Check connectivity. 'No action required. Configure appropriate link's URL. Major alarms.url.socket Error connecting to 'XXX'. YYY Major alarms.url.unknownhost Unknown host 'XXX' in url for YYY server Problem connecting to other server. Depends on reason shown. Typically network routing or security issues. Check host name is correct and host is entered in DNS and is up. Issue 5 May,

228 Alarms Severity Log File Entry English Alarm text Comment What to Do Warning alarms.vacuum It is now XXX days since the last full database vacuum. The recorder will perform a full vacuum the next time it is restarted. This may take some time. To postpone this to the following reboot, clear the setting on the System Settings > Server page. You <b>must</b> restart the recorder and allow it to vacuum the database at least once every 240 days. See text of message. Major eqconnect.reason.connectti meout Initial connection timed out with eqconnect at XXX Check URL and ensure eqconnect is up. Major eqconnect.reason.inactivityti meout Inactivity timeout with eqconnect at XXX Check eqconnect is up and network path is available. Major eqconnect.reason.ioexceptio n IO Exception with eqconnect at XXX. Reason:YYY Check network path to eqconnect. Major eqconnect.reason.linkok Successful response from eqconnect at XXX Major err.alarmtag.exception Error displaying alarm tag. Major err.alphas.invalidchar Invalid character ('-') in a non-numeric string. Major err.alphas.toolong Non-numeric list is too long character maximum including commas. Major err.archive.baddisk Bad or wrong type of archive disk inserted. Major err.archive.error Unable to write to archive file YYY. Reason: XXX. Check Troubleshooting Guide. Major err.archive.nodisk No disk available for calls archive. Major err.archive.wrongdisk An old archive disk (XXX) is inserted, but archiver needs the current disk (YYY) or a blank disk. Internal error. No action required Report problem. Correct the entry. Reduce the length of the entry. Insert the current DVD+RW or a new, blank one. Depends on reason. Insert a DVD+RW Insert the current DVD+RW or a new, blank one. 228 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

229 Alarms Table Severity Log File Entry English Alarm text Comment What to Do Major err.confighistory.recordsdelet ed One or more configuration history records have been deleted. Do not trust config history or audit records Major Err.confighistory.tampered One or more configuration history records have been altered. Major err.cscmstatetag.exception Error displaying state tag. Major err.database.purge Error purging database. Reason: XXX. Major err.eqb.1000 Licensed port count exceeded: requested XXX, available YYY Major err.eqb.1001 Insufficient ports allocated: requested XXX, allocated YYY Major err.eqb.1002 Port XXX: Invalid Parameter: YYY Major err.eqb.1003 File share XXX unreachable Major err.eqb.1004 File share XXX access denied Major err.eqb.1005 Port XXX: File I/O error on 'YYY', error: Major err.eqb.1006 Port XXX: Invalid state. Currently YYY Major err.eqb.1007 Port XXX Out of Service Internal error. Quality server tried to access more ports than are licensed. Quality server tried to access more ports than have been allocated to it. Internal Error Confirm that fileshare is reachable. Do not trust config history or audit records Report problem. Depends on reason shown. Reconfigure Quality server or increase license. Reduce port count on quality server or allocate more ports on recorder. Report problem. Open fileshare manually. Confirm access details. Confirm account has read\write access Check access to path shown. Report if persistent. Report if port is not recovered rapidly. Major err.eqb.1008 DMCC Down Check AE Server. Major err.eqb.1009 Port XXX: Local file I/O error on 'YYY', error: Major err.license.clocksetback The system clock has been set back. Timed licence is invalid. Major err.license.create Error creating licence token object Minor Err.mail.invalidaddress Invalid address: XXX Internal error. Check access to path shown. Obtain and enter a non-timed license. Report problem. Correct the address Issue 5 May,

230 Alarms Severity Log File Entry English Alarm text Comment What to Do Minor err.mail.send Error sending . Check all account entries. Send a test manually to verify settings. Major err.mail.authentication Authentication failed attempting to send . Major err.maxusagetag.exception Error displaying peak usage details. Internal error. Check user and password entries. Report problem. Major err.portpool.nocurrent No port pool specified. Internal error. Report problem. Major err.portpooltotaltag.exception Error displaying port pool totals. Major err.settingstag.exception Error in IterateSettingsTag Major err.settingstag.invalid Invalid settings group requested in IterateSettingsTag Internal error. Internal error. Internal error. Report problem. Report problem. Report problem. Major err.settingtag.exception Error in SettingTag Internal error. Report problem. Major err.settingtag.invalidfield Unrecognised field request in SettingTag Internal error. Report problem. Major err.softphonestatetag.excepti on Error showing port state details. Internal error. Report problem. Major err.stnpooltag.exception Error in StnPoolTag Internal error. Report problem. Major err.stnpooltag.invalid Invalid station pool requested in StnPoolTag Major err.stnrangetag.exception Error displaying station range details. Major err.stnrangetag.invalidfield Unrecognised field name in StnRangeTag Internal error. Internal error Internal error Report problem. Report problem. Report problem. Warning err.system.restart System restarting. No action required. Info err.system.shutdown System shut down. No action required. Major err.tags.misc Error displaying miscellaneous tag. Major err.tags.license Error displaying licence tag. Internal error Internal error Report problem. Report problem. Major err.tokentag.exception Error in TokenTag Internal error Report problem. Major err.tokentag.invalid No token set in TokenTag Major err.tokentag.invalidfield Unrecognised field name in TokenTag Internal error Internal error Report problem. Report problem. 230 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

231 Alarms Table Severity Log File Entry English Alarm text Comment What to Do Major err.usertag.exception Error in UserTag Internal error Report problem. Major errpage.label.badstuffhappen ned An error has occurred: Internal error. Report problem. Info info.archive.errorcleared Wrote to archive disk correctly. Info info.archive.rightdisk Correct archive disk now inserted. Now able to write to disk in drive No action required. No action required. Info info.archive.ewareok Eware connection OK. No action required. Info info.mail.sent Mail sent successfully. No action required Info info.queue.copy File copied successfully to XXX Info info.standby.goingstandby Standby recorder returning to idle. Info info.standby.primaryok Primary recorder requests Standby(s) go idle. Info info.standby.viable Recorder fatal error resolved. OK to record. Major link.master.linkerr Error on link to Slave recorder at XXX: YYY Major link.primary.linkerr Error on link to Standby recorder at XXX: YYY No action required No action required. No action required. No action required. Check connectivity, check other node is up. Check connectivity, check other node is up. Major standby.reason.connecttimeo ut Initial connection timed out with XXX Check connectivity, check other node is up. Major standby.reason.timeout Initial connection timed out Major standby.reason.inactivity Heartbeat failure via XXX Check connectivity, check other node is up. Check connectivity, check other node is up. Major standby.reason.noactivity Heartbeat failure Check connectivity, check other node is up. Major standby.reason.primaryreque st Instruction from MAster Recorder No action required. Major standby.reason.reconnecttim eout Connection dropped and reconnection timed out with XXX Check connectivity, check other node is up. Issue 5 May,

232 Alarms Severity Log File Entry English Alarm text Comment What to Do Major err.confighistory.tampered One or more configuration history records have been altered. Info stnrange.usage.nowarn The pool of ports XXX has YYY or more port(s) available. Minor stnrange.usage.warn The pool of ports XXX has LESS THAN YYY port(s) available. Do not trust audit records. No action required. Consider allocating more ports to this mode. 232 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

233 Appendix D: External Control Interface This appendix provides details of the external control protocol and Java class library. The main sections in this appendix are: Introduction on page 234 Java API Toolkit on page 237 TCP/IP Protocol Overview on page 238 Examples on page 242 TCP/IP Message Sequences on page 245 Issue 5 May,

234 External Control Interface Introduction External applications can control or influence recording by using the Recorder Control Protocol. You may integrate directly to a TCP/IP socket interface using the protocol described or by using the Java package provided for this purpose. An example application that uses the Java interface is also provided for reference. When to Use External Control The recorder includes sophisticated CTI interfaces and recording control but there are still occasions when additional flexibility or further data tagging is required. You will require an "external controller" in the following cases: Complex Recording Rules If your "record/do not record" requirements are more complex than the administration interface supports. For example, if you wish to start and stop recording at specific points within a call. Additional Tagging Hybrid Systems If you need to "tag" recordings with details that are not normally provided by the recorder. If your telephone system is controlled by another application (e.g. Genesys) then it may be appropriate to control recording as a result of the events occurring on a CTI feed from that application. In such cases, you may need not only the external control protocol but also Verint's Integration Framework or Unify server. These can interpret a wide range of CTI feeds. This is beyond the scope of this document. This Appendix The remaining sections of this appendix provide: a brief overview of the Java interface classes. For detailed information refer to the JavaDocs for this package. an Overview of the underlying TCP/IP control protocol 234 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

235 Introduction a reference guide for the message sequences exchanged between recorder and controller Port Allocations The Unify/External Control recording modes are provided for backwards compatibility with existing deployments only. New deployments should NOT allocate ports to these modes. Instead, use the external controller with ports in the following modes: Station Bulk - for all service observe based recording. Conferenced - for single-step conference recording where the external controller needs to provide additional tagging or is splitting long recordings into segments (as needed with many auto-diallers). Do NOT use ports in this mode if you want to initiate recordings from the external controller. You must allow the CSCM to establish conferences before sending TAG, STOP (and then subsequent START) commands. Do NOT send "START SSC:nnnn:" commands as these will conflict with the recorders allocation of ports to calls. On Demand - for single-step conference recording where the external controller is determining which calls are to be recorded. It can either instruct the recorder to establish a single-step conference; do so itself or support manual conferencing. Important:! Important: For backwards compatibility with existing deployments (in which Unify does not send the FALLBACK OFF command), the recorder automatically takes FALLBACK off those ports that are assigned to the Unify/External control (Service Observe) mode. In this case, these ports only revert to FALLBACK mode when ALL external controllers have disconnected. You can override this behaviour by setting unify.defaultrecord=true in the properties file but for new deployments should be using Station Bulk ports instead. Master + Slave Systems This topology only supports Conferenced mode recording. Hence the only permitted uses of Unify/External controllers are: Tagging calls Splitting (and tagging) long calls (e.g. autodialler sessions) with STOP and subsequent START commands. You cannot use the external controller to initiate recordings (e.g. by sending in START STN:nnn: SSC:mmm: commands) as this will conflict with the Master's port allocation algorithms. Issue 5 May,

236 External Control Interface Connect Unify/External controller to the Master (and Standby if present) only. Do NOT connect Unify to the Slaves. All ports on the slaves will be configured in On Demand Mode and the slave recorders will already be configured to consider the Master (and Standby if present) to be their "external controllers". There should be no need for the Unify/External controller to be aware that ports are on slave recorders as it communicates with the master (and standby) exactly as if the ports were on the master (or standby). The only exception to the above is that when using Quality Monitoring in "CSCM Plus" mode with external tagging, the following limitations apply to post-call tagging: Calls made before the most recent restart of the active recorder cannot be post-call tagged Only the most recent 1000 calls can be post-call tagged. (Change this default value with eqconnect.guidcache=nnnn in the properties file). Following a switchover from master to standby or vice versa, only calls started by the currently active recorder can be post-call tagged. These limitations are not normally of any concern to most applications as these tag calls very shortly after the call ends. They may affect batch mode tagging - such as overnight processing. In such cases, consider using additional table(s) in the eware database instead and updating these directly rather than tagging the recording itself. 236 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

237 Java API Toolkit Java API Toolkit Rather than write directly to the socket level interface described below, you may use the Java language API provided. This contains the following components: Packages com.swhh.cti.csapi.lang Provides a high-level API for controlling the recorder. com.swhh.cti.csapi.msg com.swhh.cti.csapi.service com.swhh.cti.csapi.tools com.swhh.cti.csapi.ui Provides Recorder Control Protocol message classes and interfaces as well as classes that implement the java.io.reader interface for reading these messages from streams. Provides a low-level API for communication with the recorder. Contains two debugging tools as well as examples of applications that use the low-level API. Contains a test harness with a graphical user interface that also serves as an example application built on top of the high-level API classes. JavaDoc for these classes is available on request. To view the JavaDoc without a Java IDE, unzip the contents and double-click index.html for a top-down view. For a comprehensive reference index, double-click indexall.html. Issue 5 May,

238 External Control Interface TCP/IP Protocol Overview Connection Method An external controller can drive the recorder via a TCP/IP socket using the protocol described below. You configure the Master recorder with the IP address and socket number of the controller(s) to which it should connect. Connection defaults to port 1414 but you can specify an alternate port. For example, you must specify port 1415 to connect to the PDS controller and port 1416 to connect a Slave recorder to a Master recorder. Note: Note: This setting was previously on the administration page for Unify/External Control ports but has been moved to the main server settings page as it now applies to all recording modes. The recorder creates a socket and attempts to connect to each specified address. If it fails to connect, it will try again every 60s. You can use multiple, independent controllers connected to a single recorder. The controlling application should open and bind a server socket and accept incoming connection requests. General Protocol Specification Messages take the form: <protocol version number> <command> (<command specific data>) '\x1b' When sending messages, follow these rules: 1. Transmit messages in UTF-8 encoding 2. All messages and responses are terminated with an escape character (27 / 0x1b). 3. Separate fields with space characters. 4. Enclose field contents in double quotation marks where the field needs to contain a space - as, for example, with filenames. Always enclose filenames in double-quotes. 5. Delimit sub-fields with colon characters ":". 238 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

239 TCP/IP Protocol Overview 6. Apart from HELLO messages (see below) and responses, all messages must be responded to in the form: <protocol version number> <error code> <channel identifier> <error qualifier> <additional info> '\x1b' Where <error qualifier> is typically "DESC:xxxxxxx:" and <additional info> may be 0 or more instances of "PARAM:pppppp:" The response to a successful command is therefore: 1 0 STN:nnn: \x1b and a typical error response is 1-1 STN:1234: DESC:"Invalid Parameter": PARAM:STN \x1b 7. Ignore unexpected or malformed responses. 8. Send commands in uppercase, but test for them case insensitively. Channel Identification In most messages below, the recorder's channels are identified by Station Number STN:nnnnn:. Where nnnnn can be 1 to 9 digits long and may include leading zeroes. The use of this station number varies according to the type of recording port being referred to: Ports using a Service Observe mode (Station Bulk and Station Executive mode) are referred to by the station number that they are observing. Ports using conferencing are referred to by the station number of the softphone that the recorder is using. XML Tagging All tagging information is passed straight through the recorder into the XML file associated with the recording. The recorder only looks for an escape character to delimit each command. The intention is that the controlling application will pass through valid XML tags which the recorder will insert into the body of the call's XML file. These may include valid Issue 5 May,

240 External Control Interface expansions of control and other characters (e.g. %20 etc). The entire xml string will be inserted between opening and closing "taglist" tags. For example, the command 1 TAG STN:1234: INUM: : <spare1>some data</spare1><spare4>john Doe</spare4>\x1b will result in call reference , which is currently being recorded on STN 1234 being tagged with the additional fields: <taglist><spare1>some data</spare1><spare4>john Doe</spare4></taglist> If you intend to use Contact Viewer as your search and replay tool (rather than the integral search and replay within the recorder) you should only use the eight special user-defined fields, labelled "spare1" to "spare8". These can be relabelled for end user viewing as part of the query design using Contact Viewer. You may use other fields but the recorder only uploads these eight "spare" fields automatically to the Viewer database. Basic Call Tagging The recorder uses DMCC's CallInformationServices interface to tag recordings in all modes except Conferenced (for which it uses Avaya CT derived information). This imposes a load on the switch. If the external controller has access to all details about the call, there may be no need for the recorder to request these details itself. You can choose whether or not the recorder collects these details itself by setting property file entries as specified in the table below: Port Type Default Override with this in prop Comment Station Bulk stnbulk.ocpneeded=false Disable only if external controller will provide all necessary details (bar date/time and duration) Station Executive On Demand Meeting stnexec.ocpneeded=false ondemand.ocpneeded=false meeting.ocpneeded=false Conferenced Not Applicable Uses Avaya CT (TSAPI) info. Unify (Service Observe) Unify (Conferenced) Disabled unify.ocpneeded=true unify.ocpneeded=true Enable if external controller does not have all call info. 240 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

241 TCP/IP Protocol Overview Port Type Default Override with this in prop Comment Quality Live Monitor Not Applicable If the recorder is collecting call details itself, the external controller can ask to be advised whenever these change. Use the REQUEST ACTIVITY ON command to do so. Fallback Mode When a recorder starts, it automatically runs in locally controlled mode - where call events it detects cause it to start/stop recording. This is also known as "fallback" mode. If the external controller only wishes to TAG calls with additional data, it may be happy to leave the ports in this mode. On the other hand, if it wishes to take control of the recording decisions then it must instruct the recorder to relinquish control of the ports it wants to use. It does this with the FALLBACK OFF command - either for a specific STN or for all ports. If the heartbeat to the external controller that turned FALLBACK off is subsequently lost, the channel(s) will revert to fallback mode and recording will continue as if the controller had never been present. Use the station specific command to take control of specific stations if you want to allow other stations to continue to be controlled by the recorder or if you are using multiple controllers, each of which manages its own set of stations. Issue 5 May,

242 External Control Interface Examples Typical examples of how to use an external controller are given below. You should read these in order as later examples refer back to earlier ones. Third-party CTI Control A bank using Genesys to route calls through an Avaya switch and wishes to record certain calls according to user defined data and events occurring within their call routing application. They would consider two designs: Service Observe Based If most - and potentially all calls at any given time - are to be recorded, a service observe approach would be appropriate: 1. Configure Station Bulk recording for all stations to be recorded. 2. The controller establishes contact with the recorder (exchanges HELLO messages) 3. The controller notes the stations that are ONLINE (and may alarm if any are missing) 4. The controller takes charge of all stations by sending 2 FALLBACK OFF 5. The controller sends START and STOP messages to control recording. You should note that a STOP message disables all recording (not just the call in progress) until a START message is sent (or the recorder falls back). 6. The recorder performs basic tagging itself but the controller sends Genesys user data through as TAG commands. Should the controller fail, the recorder will revert to normal Station Bulk mode. For a system that is tolerant of recorder failure, use a Standby recorder. This will contact the controller as and when it takes over from the primary. Single-step Conference Approach If only a small number of calls are to be recorded, it may be more cost effective to provide a pool of ports and conference these in as needed. This approach also allows supervisors to continue to use service observe. In this case: 1. Configure a pool of On Demand ports - large enough for worst case concurrent recording need. 242 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

243 Examples 2. The controller establishes contact with the recorder (exchanges HELLO messages) 3. The controller notes the pool of ports that are ONLINE (and may alarm if there are too few) 4. When a call is to be recorded, the controller picks an available port and sends it a START message with SSC parameter specifying the station that is already on that call.. 5. When recording is no longer required, the controller sends that port a HANGUP message, 6. On hanging up (whether instructed to by the controller or if the call ends) the recorder sends a HUNGUP message to the controller. Only then can the port be reused for another call. 7. The recorder performs basic tagging itself but the controller sends Genesys user data through as TAG commands. Should the controller fail, no recording will happen. For fault tolerance, use a pair of controllers and have one or the other send messages to the recorder. For large systems or to make the system tolerant to recorder failure, use a pool of N+1 independent recorders. Additional Call Tagging User wants to tag each call with a customer identifier that is known to an in-house application. This knows the station handling the call and the customer on the call in real-time i.e. during the call. In this case, the recorder will be configured for either Station Bulk or Conferenced mode as appropriate for the basic recording requirement. The approach to tagging will differ slightly. Service Observe Approach If Station Bulk recording is used: 1. Configure Station Bulk recording for all stations to be recorded. 2. The controller establishes contact with the recorder (exchanges HELLO messages) 3. The controller notes the stations that are ONLINE (and may alarm if any are missing) 4. Any time the controller receives a STARTED message, it looks up its current map of which stations v customers and sends a TAG command to apply the appropriate customer identifier to the call. Should the controller fail, the recorder continues to record as normal in Station Bulk mode but calls will not be tagged with customer id. A fault tolerant pair of controllers can be used if required. Issue 5 May,

244 External Control Interface For a system that is tolerant of recorder failure, use a Standby recorder. This will contact the controller as and when it takes over from the primary. Single-step Conference Approach If Conferenced mode is used: 1. Configure Conferenced mode as normal. 2. The controller establishes contact with the recorder (exchanges HELLO messages) 3. The controller notes the recorder ports that are ONLINE (and may alarm if any are missing) and requests ACTIVITY ON for each. 4. Any time the controller receives an UPDATE message, it looks in the XML data for a station identifier that it recognises. If it finds one, it sends a TAG message with the appropriate customer identifier. (CAUTION: this will, in turn cause another UPDATE message to come back. Do not go recursive! If the UPDATE message already includes a customer identifier, do not send TAG again.) 5. Should a recorder port hang up, it will send "1 HUNGUP STN:nnnn:". The external controller should use this message to indicate that the port is once again available for use. Should the controller fail, the recorder continues to record as normal in Conferenced mode but calls will not be tagged with customer id. A fault tolerant pair of controllers can be used if required. For a system that is tolerant of recorder failure, use a Standby recorder. This will contact the controller as and when it takes over from the primary. 244 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

245 TCP/IP Message Sequences TCP/IP Message Sequences The table below shows all supported message sequences: Controlling Application Sends Dirn Recorder Sends Comments On recorder first establishing connection with controller s socket (default 1415) <<< 1 HELLO CMAPICapture NNNNNN The controller can determine which recorder is which from the port that it connects on, its IP address or its serial number NNNNNN. When using Unify, the former method is normally used and hence each recorder should be configured to contact a different port number. 1 HELLO XXXXX >>> Where XXXXX is the application s name or other identifier. The recorder does not parse beyond HELLO. The recorder writes the full message to the log file but does not act on the name given. To take control of recording on all ports <<< 1 ONLINE STN:nnnnn: For each softphone configured and available, the recorder sends this message. This allows the controller to build a table of available stations and respond with 1 FALLBACK OFF STN:nnnnn: if it wishes to take control of that port. 2 FALLBACK OFF >>> External controller must now enable/disable recording on this station. To relinquish control of recording on all ports <<< 1 0 Confirmed 2 FALLBACK ON >>> Recorder now controls stop/start on all ports in all modes To take control of recording on a specified station 2 FALLBACK OFF STN:nnnn: <<< 1 0 Confirmed >>> External controller must now enable/disable recording on this station. Issue 5 May,

246 External Control Interface Controlling Application Sends Dirn Recorder Sends Comments <<< 1 0 STN:nnnnn: Confirmed To relinquish control of recording on a specified station 2 FALLBACK ON STN:nnnn: To activate additional notifications 1 REQUEST ACTIVITY ON STN:nnnn: >>> Recorder now controls stop/start on this station. <<< 1 0 STN:nnnnn: Confirmed >>> The controller can ask that it is advised of call details in addition to basic started/stopped messages. <<< 1 0 STN:nnnnn: Confirmed Start Recording (optionally, set up single-step conference) 1 START STN:nnnnn: SSC:mmmm: <xml data> Note: SSC parameter is optional. Stop Recording 1 STOP STN:nnnnn: <xml data> >>> Start recording on station specified (nnnn) and tag with optional data supplied. If SSC parameter is provided, the recorder will use TSAPI to request that the recorder s softphone (nnnnn) be single-step conferenced onto the call currently active on station (not agent or VDN) mmmm. Use this parameter with OnDemand, Conferenced and Unify (Conferenced) ports only. Failure to attempt to establish a connection (e.g. because the call terminated before connection could be established) is indicated by an error message with NOSSC in the description. Parameter 1 repeats the target address and Parameter 2 gives the error message. <<< 1 0 STN:nnnnn: Confirmed >>> Stop recording on station specified and tag with optional data supplied. Note: This command does not cause softphone to hang up. <<< 1 0 STN:nnnnn: Confirmed External controller wishes to terminate single-step conference recording on known softphone 246 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

247 TCP/IP Message Sequences Controlling Application Sends Dirn Recorder Sends Comments 1 HANGUP STN:NNNN: <optional XML data> On recording starting On recording stopping On recording port hanging up >>> Note that the STOP command simply stops recording it does not cause the port to hang up. Hence an additional command is required where a port is conferenced into a call.: This command instructs the specified station to hang up implicitly terminating any current recording and, optionally, tagging it with the XML data provided. However, the recording state (enabled/disabled) of the softphone is NOT changed. <<< 1 0 STN:nnnnn: Confirmed <<< 1 STARTED STN:nnnn: INUM:nnnnnnnnn nnnnnn: <<< 1 STOPPED STN:nnnn: INUM:nnnnnnnnn nnnnnn: <<< 1 HUNGUP STN:nnnn: Unify can note the INum to ensure subsequent tag commands go to the right segment even if that call has finished. Unify can also parse out the station number if it wants to check for orphaned records hanging around from this station which are obviously now in error. We could drop the call if told to but we can t be sure it won t stop before then so maybe best to always report when call drops. Don t see any need to include tagging as we will not normally be getting any additional tagging beyond stn number. On recording getting new or changed info (NAILED mode and only if ACTIVITY ON previously requested for this port) On Unify wanting to tag call <<< 1 UPDATE STN:nnnn: INUM:nnnnnnnnn nnnnnn: <xml data> Info available for call specified on station specified. Unify may need to interpret some of the xml to determine recording action. Will send all call info on each update so Unify can cache single string. Issue 5 May,

248 External Control Interface Controlling Application Sends Dirn Recorder Sends Comments 1 TAG STN:nnnn: INUM:nnnnnnnnnnnnnnn: <xml data> Call Break Unify initiated 1 BREAK STN:nnnn: <xml data> >>> This allows tagging to be sent for the current and previous calls on a channel. The recorder will use nnnnnnnnnnnnnnn to determine whether the tagging is for the current call or the previous one. In the latter case, it will open the XML file and append the details given. Additional tagging should currently be confined to the 8 pre-defined fields spare1 to spare8. i.e. <xml data> should be of the form <spare1>some data</spare1>. <<< 1 0 STN:nnnnn: Confirmed >>> Recorder breaks the call seamlessly. The optional tag data is applied to the stopped call. Send further TAG command to tag new call. Noend and nostart fields are set true on the old and new calls as per MediaStore. <<< 1 0 STN:nnnnn: Confirmed <<< 1 STOPPED STN:nnnn: INUM:nnnnnnnnn nnnnnn: <<< 1 STARTED STN:nnnn: INUM:nnnnnnnnn nnnnnn: Call Break at recorder due to max segment duration exceeded <<< 1 STOPPED STN:nnnn: INUM:nnnnnnnnn nnnnnn: <<< 1 STARTED STN:nnnn: INUM:nnnnnnnnn nnnnnn: Same STN but different INum Same STN but different INum 248 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

249 TCP/IP Message Sequences Controlling Application Sends Dirn Recorder Sends Comments Call Masking <<< 1 UPDATE STN:nnnn: INUM:nnnnnnnnn nnnnnn: <xml data> Only sent if ACTIVITY ON previously requested. Unify can identify this as a broken call from the nostart field being true. Unify tagged data will have been rolled over from previous call and included in this call s data already. Note: The properties file setting rec.maskallowed must be set to true for this feature to be enabled. 3 PAUSE STN:nnnn: >>> Instructs the recorder to begin masking. The recorder continues to record, but replaces the real audio with a fixed tone pattern. This is useful for portions of conversations that are confidential. <<< 3 PAUSED STN:nnnn: 3 RESUME STN:nnnn: >>> Instructs the recorder to revert to normal recording of the real audio. On shutdown <<< 3 RESUMED STN:nnnn: 1 BYE <reason> >>> <reason> can be 1. FALLBACK recorder should continue to answer calls and record (though in most cases, no calls will appear unless Unify is sending them to the recorder s stations). Without knowing the mappings in Unify, it is not as easy to drop into a fallback mode unless station numbers are also configured somewhere. 2. SHUTDOWN recorder will stop recording on Unify controlled ports. Allows busying out. On channel error <<< 1 BYE <extra info> message Issue 5 May,

250 External Control Interface Controlling Application Sends Dirn Recorder Sends Comments On channel being restored <<< 1 ERROR STN:nnnn: <<< 1 ONLINE STN:nnnn: Softphone unusable e.g. cannot register. In AUTO-ANSWER mode, Unify should mark this port faulty and not attempt to use. Hopefully, an ONLINE message will follow shortly as the recorder re-registers the port. Unify should probably be written to recognise faulty behaviour too e.g. if it doesn t get STARTED from softphone it tried to place a recording on, it needs to report this and busy out the faulty one? Softphone available 250 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

251 Appendix E: Fault Tolerant Systems In addition to using fault tolerant components within servers as described in High Availability Systems on page 59, recording systems can be made tolerant of many server and network failure conditions. This appendix details how such systems are designed and configured, how they handle failures and how to upgrade them. This Appendix assumes that the reader is already familiar with design, installation and operation of single server recording systems as described in the body of this manual. The main sections in this appendix are: Redundant SAN on page 252 Duplicated recording on page 253 Standby Recorder Options on page 254 Mode of operation on page 276 Standby Recorder Configuration on page 281 Upgrade Process for Primary/Standby on page 283 Issue 5 May,

252 Fault Tolerant Systems Redundant SAN If you already have a fault tolerant storage network in place, you can assign the calls storage path to an area of this storage system. This is supported for locally connected drives - NOT for Network Attached Storage (NAS). You are then reliant on the mirroring or other redundancy and standby mechanisms associated with the storage network. These mechanisms typically include a tape library backup and might include hierarchical file storage (HFS), in which older files are replaced by small tokens that allow the system to retrieve the original content from the tape library. The recorder runs successfully with Tivoli Storage Manager and might also support other similar systems. However, Verint does not proactively test against these systems. Connect to them at your own risk. If you want to use such a system, turn the recorder's own disk management function off. To do this, add the line disk.manager=false to the properties file. This stops the recorder from deleting the oldest files as the available disk capacity falls to 1GB. You must then ensure that the available space on the /calls partition does not fall below 1GB. 252 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

253 Duplicated recording Duplicated recording Where recording is controlled by an external application (such as Unify) establishing conference calls onto the recorder's ports, a fault tolerant system can be deployed by using two totally independent recorders and establishing 4-way, instead of 3-way conferences. By conferencing in one port from each of the two recorders, a fully redundant recording system can be delivered.! WARNING: WARNING: Although the recorders are running in parallel, you must also ensure that the mechanism that is establishing the conferences is, itself, fault tolerant. This may imply duplicated Unify servers and/or duplicated CTI feeds. The default behavior of recorders is to fall back into basic recording mode (where possible) should connectivity to Unify be broken. This often suffices as an acceptable failure mode. You must ensure that the mechanism that establishes the conferences can handle the error condition that occurs when one of the recorders fails. It must establish the other conference party to the remaining, good recorder rather than fail totally. Note: WARNING: Note: A parallel recording system inevitably requires twice the resources on the Communication Manager, both in terms of IP_API_A licenses and VoIP resources. This option is, therefore, an expensive one.! WARNING: The overall maximum of 6 parties on a conference might become a constraint if two of these are recording ports. Call scenarios in which two calls merge (transfer and conferencing) - and where each of these is already a 4 party call due to duplicated recording can exceed these limits and are not supported. To avoid this, recording ports are removed from a call as soon as it goes on hold. In this way there are less parties present on the call when it later merges with a consultation call. Issue 5 May,

254 Fault Tolerant Systems Standby Recorder Options You can configure additional Verint Contact Recorders to act as standby to a primary recorder. These would normally be in "standby" mode and only become "active" when they detect the failure of the primary they are shadowing. Where you have multiple sites, you can use multiple standby recorders. Each standby can back up a specified primary or a specific subset of a primary recorder's configuration. This section describes: Prerequisites for high availability Standby recorder licensing Known limitations Supported fall back scenarios Supported failure modes Mode of operation The impact of a switch-over Restoration of a primary recorder once it has been repaired Upgrade procedures specific to primary / standby topologies A summary comparison with hardware switch-over units "Supported Master/Standby Topologies Prerequisites for high availability To get maximum benefit from a standby recorder topology, you must adopt the appropriate elements of an overall strategy for high availability: Centralized database and replay Although the standby unit(s) can work without a centralized database and replay tools like Viewer, you need these elements to provide a seamless, single point of access to all recordings. Without them, those recordings made on a standby recorder are only accessible by accessing the url of that recorder. With the central database, all recordings made on any of the recorders, including the standby(s), are consolidated into a single, searchable pool. 254 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

255 Standby Recorder Options Fault tolerant storage Off-site archive Both primary and standby recorders must be configured with RAID1 or (preferably) RAID 5 arrays or be using a fault tolerant SAN on which recordings are kept. Using standby recording provides fault tolerant recording. However, it does not itself provide for fault tolerant STORAGE of recordings. It is also assumed that, where DVD+RW disks are written, that these will be stored away from the recorders, ideally in a separate building for maximum protection. Should these be needed for regular access, a copy should be made and kept off-site. Unless the fault tolerant disk system is spread across multiple buildings, off-site storage of archive media is essential. Fault tolerant network connectivity Where centrally located primary and standby recorders are co-located or are in disaster recovery sites with fault tolerant connections between them, it is critical that the primary and standby recorders are connected by way of a fully fault tolerant network topology. Failure of any one networking component (NIC, cable, Ethernet switch or router) between them must not result in connectivity being lost between them. Each standby recorder assumes that failure to communicate with the primary indicates that the primary has failed. It will attempt to take over from it. If this is not the case, then the two recorders will clash as they attempt to use the same softphone numbers on the switch. You must provide multiple independent paths or teamed NICs with independent network paths between primary and standby as described in Fault Tolerant Paths on page 84. Where remotely sited standby recorders are provided so as to record should the remote site lose connectivity with the center and fall back to ESS mode, you should ensure that the remote standby is connected to the primary recorder over the same network path that the Avaya components use. This ensures the it will lose contact with the center - and hence become active - should the site fall into ESS mode. Location of recorders Where possible, you should locate the standby recorder(s) in a different building from the primary recorders. You should also provide diverse network routing between the recorders and the Avaya switch components that they interact with (AE Services, C-LAN and MedPros). One additional Media Processing Resource per Media Gateway You must provide one additional media processor per Media Gateway to ensure that failure of a single card does not impact the ability of a recorder to record at its configured capacity. Issue 5 May,

256 Fault Tolerant Systems Standby recorder licensing The standby recorder is given a special server license that: modifies its Administration User Interface to support the entry of the addresses of the recorder it is to back up. restricts it from being used as a primary recorder. Once the Standby Server is licensed and its primary identified, the standby copies the channel licensing details from the Primary. However, when the standby recorder is manually or locally configured, it does not copy channel license details from the primary. In this circumstance, you must ensure that it is licensed appropriately. When you enter a license key that identifies a recorder as a standby unit, you must subsequently stop and restart the recorder in order for it to take notice of this setting. In laboratory conditions, when switching between primary and standby licenses, a restart is required. Automated Configuration Shadowing If you are using service observe based recording (Station Bulk, Station Executive) you will probably want the standby recorder on each site to start recording those stations that are on that site when failover occurs. Remote standby recorders copy the entire configuration from the central primary recorder. In the event of network failure, they start recording those stations that are marked with that standby recorder's serial number as their standby recorder. They will do so using softphones that are marked with their serial number as their Standby recorder. These settings are under the "Advanced" link for the range of softphones or stations. Manual Configuration of Standby If you are using conferenced mode recording, it is unlikely that your fallback scenarios will include full CTI links. In such cases you can instruct the Standby recorder NOT to copy configuration from the primary. You are then free to configure it manually such that it operates as you require under fallback conditions. This may be a simpler approach such as recording all stations on its site. See Standby Server on page 182 for instructions on how to specify this option. 256 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

257 Standby Recorder Options Known limitations Deploying standby recorders: Does not guarantee access to recordings made by the failed server. See Centralized database and replay above. Does not guarantee "no loss" of recording on failure. See Mode of operation below. Requires that the standby recorder can contact the primary recorder when it is first configured so that its configuration can be copied. Does not allow configuration changes to configuration while a standby recorder is active (unless the standby is configured for local configuration). You must restore the primary recorder before making any changes to the configuration. Does not support the use of Telephone Replay ports on the primary and standby recorder (unless you force local configuration). Place these on a central replay server (or pair of servers if fault tolerant replay is required). Distributed Systems The primary and standby operations are designed to work well in distributed systems where loss of connectivity between sites would otherwise case recording to fail. The following Avaya failover modes are supported: Locally Survivable Processors (LSP) Enterprise Survivable Servers (ESS) AE Server to Recorder Mapping Each group of primary recorders and each group of standby recorders should have their own AE Services box. This provides an extra layer of redundancy. If both the primary and standby share the same AE Server then a failure of this server will affect both. If they connect to different AE Servers then the primary recorder can detect this and ask the standby to take over. As it uses a different AE server, it should be able to do so successfully. Where recording is taking place across the enterprise, always place an AE server on the same site as the recorder that will use it so that the remote recorders can continue to operate should the WAN fail. Locally Survivable Processors (LSP) LSP is where a remote site has an S8300 chassis that normally acts as a slave to a central S8700. When the connection to the central site fails, the S8300 takes over in LSP mode Issue 5 May,

258 Fault Tolerant Systems and uses a copy of the configuration from the central switch server. Because the copied switch configuration (from the central site) cannot show the S8300 itself offering a DAPI connection, the switch will not allow the recorder to establish a DAPI link to it under fallback conditions. Therefore, recordings that are made under these conditions have limited call details - that is, they do not have data in the agent, VDN, or UCID fields. You must: specify the S8300 on the LSP site as the backup C-LAN for the recorder ports that you wish to use when in LSP failover mode specify the appropriate remote Verint Contact Recorder as the Standby recorder for the stations to be recorded there in LSP mode. Three options are supported for LSP: a. Standalone Recorder(s) on LSP Site(s) - In this enterprise configuration, the recorders and AE servers are based on the satellite site(s) only. There are no standby recorders. Each Verint Contact Recorder is configured to register its softphones through a C-LAN at the central site. The backup C-LAN address for the softphones is the S8300 on the same site as the recorder. In normal operation, the recorder uses the C-LAN on the S8700. However, when LSP mode is invoked, the recorder will reregister softphones on the backup C-LAN - in this case, the S8300 on the remote site. All calls which are taking place at the time will be lost but all new calls will be recorded by the standby. b. Automatic Standby Recorder(s) on LSP Site(s) - In this enterprise configuration, the primary recorders are based at a central site and the standby recorders are based at the satellite sites. The recording ports on the primary recorder are split into groups. Each group of ports reflects the ports on a standby recorder that will take over in failure mode. A group of ports is configured with: - The serial number of the standby Recorder that is to back these up in failover mode - Their C-LAN setting configured to point first to a C-LAN at the central site (primary) and secondly, at a C-LAN on the satellite site (backup). The port allocations made on the primary recorder are similarly marked with the serial number of the standby recorder. The standby takes over this part of the total recording load in failover mode. The configuration of the primary recorder is copied periodically to all the standby recorders on the remote sites as normal. When a failure occurs, the standby recorder connects only those ports marked with its serial number to the standby C-LAN and allocate them to the recording tasks that are marked for this recorder to perform in failover mode. c. Manually Configured Standby Recorder(s) on LSP site(s) - The system topology is the same as in b) above, with the primary recorder(s) based at a central site and 258 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

259 Standby Recorder Options the standby recorders based at the local sites. However, the standby recorders are setup for manual configuration (by means of a property file entry) and are typically configured to perform station bulk recording of the stations at the LSP site. To set this up, the primary recorder is configured without the backup or standby settings. Then, each standby recorder is configured manually to operate as required when LSP mode is invoked. Enterprise Survivable Servers (ESS) Enterprise Survivable Servers are where one or more S8500-based Communication Manager servers can take control of a portion of the overall system if the main server fails. If an S8500 is present on a remote site then it can take over the running of any gateways on that site including the C-LANs on the gateway. This is better than LSP mode from a recording perspective as this can offer a DAPI link to the standby recorder even when operating in failover mode i.e. disconnected from the central site. Two options are supported: 1. Standalone Recorder(s) and AE Server(s) on ESS Site(s) - In this configuration, the recorders are based on the S8500 site(s) only. There are no standby recorders. Here the recorders are configured to use a C-LAN on their site and, after ESS is invoked, this same C-LAN is still present and usable. The recorder may encounter a short interruption in service but it continues to function after the local S8500 has switched in. You do not need to setup the recorder for this to take place, as, apart from the brief interruption of service, the recorder is not aware of any changes. 2. Primary Recorder centrally + Standby Recorder(s) on ESS Site(s): - If the WAN is lost, the standby recorders determine that the primary is down. When they no longer detect the heartbeat signal over the WAN, the standby recorders take over. They operate in an identical manner to the standby recorders in LSP mode as described above for both the manual and local standby LSP mode. However, unlike LSP mode, their AE Server establishes a DAPI link and provides full call tagging. Supported failure modes The Standby recorder is designed to become active within a few seconds of a complete system failure of the primary recorder it is configured to "shadow." These failure modes include: Catastrophic server failure Total network isolation Accidental power down or power failure Sustained failure of the link to AE Services or (where used) Avaya CT Disk full Issue 5 May,

260 Fault Tolerant Systems Standby recorders and Unify/External Control Unify/External Controller connected to Master/Standby pair When using Unify or an equivalent external controller with a primary/standby recorder pair, it should send any START/STOP commands to the recorder that sent it port ONLINE messages. Only the active recorder will send STARTED/STOPPED messages. Unify should only send TAG commands to this active recorder. Unify/External Controller connected to Master only If the external controller is essential to the operation of the Master (for example, if it is setting up single step conferences) then the master needs to recognize that it is not viable if it has no links to external controllers. Set the property unify.required in the properties file to indicate this. You can then provide either: a standby recorder, locally configured for Station Bulk mode a second Unify/External controller connected to the standby recorder With this property set, the Unify/External controller can force a switchover to the standby by sending a 1 BYE SHUTDOWN message and then not sending the HELLO response when the recorder re-establishes connection. When it is ready to take control again, it can send the HELLO message and the recorder will take over from the standby again. Supported Master/Standby Topologies The following pages describe the supported Master/Standby topologies. Use the table below to determine the appropriate configuration for your system then refer to that table for details of sizing, configuration, how faults are handled and how to restore the system following a failure. For further discussion of these options, read the remainder of this Appendix. 260 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

261 Standby Recorder Options Recording Mode System Size Avaya Topology See Table Station Bulk Recording Single Recorder Single Site One or more remote sites with ESS support. No backup for stations recorded at central site. A B One or more remote sites with ESS support. Backup for stations recorded at central site. C Multiple Independent Recorders Single Site D Conferenced Recording Single Recorder Single Site One or more remote sites with ESS support. No backup for stations recorded at central site. A E Multiple Recorders Single Site One or more remote sites with ESS support. Backup for stations recorded at central site. F G Master/Standby Topology A If your System Type is "A" To protect against Recording Mode System Size Avaya Topology Recorder AE Server Network Station Bulk Recording or Conferenced Recording Single Recorder Single Site Critical Failure Critical Failure Failure of path between recorder and AE Server Issue 5 May,

262 Fault Tolerant Systems Master/Standby Topology A Provide Configure In the event of a failure Standby Recorder Standby AE Server Network Connectivity Replay Server Master Recorder Standby Recorder When switchover occurs Until problem is resolved When problem is resolved CPU size as per Master recorder. Disk may be smaller as long as you restore the master promptly following any switchover. Identical to other AE Server Fault tolerant connection between master and standby recorder. (Teamed NICs and independent switches or two completely independent paths.) Provide eware/viewer or Central Replay Server if you want to avoid having to search separately across Master and Standby recorders. To connect to one AES only To connect to other AES only. To connect to Master via fault tolerant path as above. Other recording configuration will be copied automatically from the Master recorder. It may take a few minutes for the switch to release existing softphone registrations and associated service observes. Calls in progress will be recorded as soon as the softphones on the standby can be registered and service observe is allowed. Standby will record exactly as the Master would have. Master will allow Standby to continue recording until you either restart the whole system or instruct it to restore via url /servlet/cscm?cmd=restore. Do this out of hours as there will be a brief interruption to recording as the system switches back. 262 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

263 Standby Recorder Options Master/Standby Topology B If your System Type is "B" To protect against Provide Recording Mode System Size Avaya Topology Recorder AE Server Network Standby Recorder(s) Standby AE Server Network Connectivity Replay Server Station Bulk Recording Single Recorder One or more remote sites with ESS support. No backup for stations at central site. Critical Failure Critical Failure Failure of path between recorder and AE Server Failure of path between an ESS site and central recorder For each ESS site you want to protect, provide a standby with CPU sized according to the number of channels to be recorded on that site. Disk may be smaller as long as you restore the master promptly following any switchover. On each ESS site you want to protect Network path between each remote standby recorder and the central recorder. Use same path(s) as are used for the Avaya switch so that if the site goes into ESS mode, the standby will also have lost connectivity to the (central) master recorder. Provide eware/viewer or Central Replay Server if you want to avoid having to search separately across Master and Standby recorders. Issue 5 May,

264 Fault Tolerant Systems Master/Standby Topology B Configure Master Recorder Each Remote Standby Recorder To connect to central AES only. Split the softphone ranges on the Settings > Communication Manager page into groups suitable for use by each standby. Use the Advanced Settings link to specify which Standby recorder should use each block of softphones. Set appropriate Standby recorder serial number (under Advanced settings link) for each station or range of stations to be recorded in Station Bulk mode (unless using localconfig option below) Each to connect to AES on its site only. To connect to Master via network path as above. Other recording configuration will be copied automatically from the Master recorder and each standby recorder will record only those stations set with its serial number as their designated standby recorder. ALTERNATIVELY set standby.localconfig=true and configure the standby itself if you want it to record different stations when in ESS mode. Ensure each standby has its own set of softphones in this case. 264 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

265 Standby Recorder Options Master/Standby Topology B In the event of a failure When switchover occurs Until problem is resolved When problem is resolved If the master recorder has failed, it may take a few minutes for the switch to release existing softphone registrations and associated service observes. Calls in progress will be recorded as soon as the softphones on the standby can be registered and service observe is allowed. Standby will record only those stations marked with its serial number as their standby recorder (OR if using the localconfig option above, it will record whatever it has been configured to record). Master will attempt to take control as soon as it recovers. This is to ensure that any stations without a remote standby assigned are recorded. Remote sites that are in ESS mode will remain in ESS mode until manually reset so will not be affected by this. Standby will continue to record until reset. You must therefore reset the standby at the same time as you restore the Avaya system from ESS mode to centralized control. Do this out of hours as there will be a brief interruption to recording as the system switches back. Master/Standby Topology C If your System Type is "C" To protect against Recording Mode System Size Avaya Topology Recorder AE Server Network Station Bulk Recording Single Recorder One or more remote sites with ESS support AND backup needed for stations at central site being recorded. Critical Failure Critical Failure Failure of path between recorder and AE Server Failure of path between an ESS site and central recorder Issue 5 May,

266 Fault Tolerant Systems Master/Standby Topology C Provide Standby Recorders Standby AE Server Network Connectivity Replay Server At Central Site, provide a Standby recorder with same CPU size as the Master. For each ESS site you want to protect, provide a standby with CPU sized according to the number of channels to be recorded on that site. Disks on all Standbies may be smaller as long as you restore the master promptly following any switchover. Identical to existing AES at central site. One on each ESS site you want to protect Fault tolerant connection between Master and central Standby recorder. (Teamed NICs and independent switches or two completely independent paths.) Network path between each remote standby recorder and the central recorder. Use same path(s) as are used for the Avaya switch so that if the site goes into ESS mode, the standby will also have lost connectivity to the (central) master recorder. Provide eware/viewer or Central Replay Server if you want to avoid having to search separately across Master and Standby recorders. 266 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

267 Standby Recorder Options Master/Standby Topology C Configure Master Recorder Central Standby Recorder Each Remote Standby Recorder To connect to one central AES only. Split the softphone ranges on the Settings > Communication Manager page into groups suitable for use by each standby. Use the Advanced Settings link to specify which Standby recorder should use each block of softphones. Set appropriate Standby recorder serial number (under Advanced settings link) for each station or range of stations to be recorded in Station Bulk mode (unless using localconfig option below) Set standby.main=true To connect to the other central AES (not the one the master connects to). To connect to Master via fault tolerant path as above. Other recording configuration will be copied automatically from the Master recorder. Each to connect to AES on its site only. To connect to Master and Central Standby via same network path Avaya system relies on for central control. Other recording configuration will be copied automatically from the Master recorder and each standby recorder will record only those stations set with its serial number as their designated standby recorder. ALTERNATIVELY set standby.localconfig=true and configure the standby itself if you want it to record different stations when in ESS mode. Ensure each standby has its own set of softphones in this case. Issue 5 May,

268 Fault Tolerant Systems Master/Standby Topology C In the event of a failure If Master recorder fails When switchover to ESS occurs Until problem is resolved When problem is resolved Central standby will take over. Remote standbys will stay inactive. It may take a few minutes for the switch to release existing softphone registrations and associated service observes. Calls in progress will be recorded as soon as the softphones on the central standby can be registered and service observe is allowed. The affected standby will register the appropriate softphones and start to record the stations assigned to it (or locally configured if that option has been set). Appropriate standby will record as above. If the Master recorder failed, DO NOT RESTART IT UNTIL OUT OF HOURS. STOP THE CENTRAL STANDBY BEFORE RESTARTING THE MASTER. Failure to do so will result in the remote standbies being told to go active. If the problem was with network connectivity, then the remote site will have dropped into ESS mode and the master will not be able to take control until this is restored as below. Remote Standby will continue to record until reset. You must therefore reset the remote standby at the same time as you restore the Avaya system from ESS mode to centralized control. Do this out of hours as there will be a brief interruption to recording as the system switches back. Master/Standby Topology D If your System Type is "D" Recording Mode System Size Avaya Topology Station Bulk Recording Multiple Independent Recorders Single Site 268 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

269 Standby Recorder Options Master/Standby Topology D To protect against Provide Configure In the event of a failure Recorder AE Server Network Standby Recorders (one for each master recorder) Standby AE Server Network Connectivity Replay Server Each Master Recorder Each Standby Recorder When switchover occurs Until problem is resolved When problem is resolved Critical Failure of any Master Recorder Critical Failure Failure of path between recorder and AE Server CPU size for each as per the Master recorder it is backing up. Disk may be smaller as long as you restore the affected master promptly following any switchover. Identical to other AE Server (provide more on very large systems) Fault tolerant connection between each master and its standby recorder. (Teamed NICs and independent switches or two completely independent paths.) Should already be in place. No need to change To connect to one AES only. Spread masters across the available AE Servers. To connect to one AES only - not the same one as that of the Master recorder it is shadowing. To connect to the one Master it is shadowing via fault tolerant path as above. Other recording configuration will be copied automatically from the appropriate Master recorder. It may take a few minutes for the switch to release existing softphone registrations and associated service observes. Calls in progress will be recorded as soon as the softphones on the standby can be registered and service observe is allowed. Each Standby will record exactly as the Master it is shadowing would have. Master will allow its Standby to continue recording until you either restart both Master and its Standby or instruct it to restore via url /servlet/cscm?cmd=restore. Do this out of hours as there will be a brief interruption to recording as this pair of recorders switches back. Issue 5 May,

270 Fault Tolerant Systems Master/Standby Topology E If your System Type is "E" To protect against Provide Configure Recording Mode System Size Avaya Topology Recorder AE Server Network Standby Recorder(s) Standby AE Server Network Connectivity Replay Server Master Recorder Each Remote Standby Recorder Conferenced Recording Single Recorder One or more remote sites with ESS support. No backup for stations at central site. Critical Failure Critical Failure Failure of path between recorder and AE Server Failure of path between an ESS site and central recorder For each ESS site you want to protect, provide a standby with CPU sized according to the number of channels to be recorded on that site. Disk may be smaller as long as you restore the master promptly following any switchover. On each ESS site you want to protect Network path between each remote standby recorder and the central recorder. Use same path(s) as are used for the Avaya switch so that if the site goes into ESS mode, the standby will also have lost connectivity to the (central) master recorder. Provide eware/viewer or Central Replay Server if you want to avoid having to search separately across Master and Standby recorders. To connect to central AES only. Each to connect to AES on its site only. To connect to Master via network path as above. Set standby.localconfig=true and configure the standby to record the local stations. You may want to have the standby record in Station Bulk mode rather than rely on local TSAPI services. 270 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

271 Standby Recorder Options Master/Standby Topology E In the event of a failure When switchover occurs Until problem is resolved When problem is resolved Calls in progress will be recorded as soon as the softphones on the standby can be registered and service observe is allowed. Standby will record whatever it has been configured to record. Standby will continue to record until reset. You must therefore reset the standby at the same time as you restore the Avaya system from ESS mode to centralized control. Do this out of hours as there will be a brief interruption to recording as the system switches back. Master/Standby Topology F If your System Type is To protect against Provide Recording Mode System Size Avaya Topology Recorder AE Server Network Standby Recorder Standby AE Server Slave Recorders Network Connectivity Replay Server Conferenced Recording Multiple Recorders Single Site Critical Failure Critical Failure Failure of path between recorder and AE Server CPU size as per Master recorder. Disk of Master and Standby do not record calls so may be small. Identical to other AE Server Provide one more slave recorder than is needed to support maximum concurrent load. These operate as a pool so N+1 are required for fault tolerance. Fault tolerant connection between master and standby recorder. (Teamed NICs and independent switches or two completely independent paths.) Should be already provided. Issue 5 May,

272 Fault Tolerant Systems Master/Standby Topology F Configure In the event of a failure Master Recorder Standby Recorder Slave Recorders When switchover occurs Until problem is resolved When problem is resolved To connect to one AES only. ZERO concurrent recording channels in Conferenced page (all recording is done on slaves). To connect to other AES only. ZERO concurrent recording channels in Conferenced page (all recording is done on slaves) To connect to Master via fault tolerant path as above. Other recording configuration will be copied automatically from the Master recorder. To connect to Master and Standby recorders as their "external controllers". All ports in On Demand mode. Do NOT configure Conferenced mode or any Avaya CT settings. Apart from a few seconds delay, calls in progress should be unaffected if the Master fails. If a slave fails, the calls being recorded on it will be reassigned to the remaining slaves within a few seconds. Standby will control recordings exactly as the Master would have. Master will allow Standby to continue recording until you either restart the whole system or instruct it to restore via url "/servlet/cscm?cmd=restore". Do this out of hours as there will be a brief interruption to recording as the system switches back. A failed slave, on the other hand, can restored without danger. Note that its ports will be added to the back of the available port list so it may be some time before calls are assigned to it. 272 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

273 Standby Recorder Options Master/Standby Topology G If your System Type is "G" To protect against Provide Recording Mode System Size Avaya Topology Recorder AE Server Network Central Standby Recorder Remote Standby Recorder(s) Standby AE Server Slave Recorders Network Connectivity Replay Server Conferenced Recording Multiple Recorders One or more remote sites with ESS support AND backup needed for stations at central site being recorded. Critical Failure Critical Failure Failure of path between recorder and AE Server CPU size as per Master recorder. Disk of Master and Standby do not record calls so may be small. For each ESS site you want to protect, provide a standby with CPU sized according to the number of channels to be recorded on that site. Disk may be smaller as long as you restore the system promptly following any switchover. One on central site identical to existing AE Server AND one on each ESS site you want to protect. Provide one more slave recorder than is needed to support maximum concurrent load. These operate as a pool so N+1 are required for fault tolerance. Fault tolerant connection between master and standby recorder. (Teamed NICs and independent switches or two completely independent paths.) Network path between each remote standby recorder and the central recorders. Use same path(s) as are used for the Avaya switch so that if the site goes into ESS mode, the standby there will also have lost connectivity to the central recorders. Should be already provided. Issue 5 May,

274 Fault Tolerant Systems Master/Standby Topology G Configure Master Recorder Central Standby Recorder Remote Standby Recorder(s) Slave Recorders To connect to one AES only. ZERO concurrent recording channels in Conferenced page (all recording is done on slaves). To connect to other AES only. ZERO concurrent recording channels in Conferenced page (all recording is done on slaves) To connect to Master via fault tolerant path as above. Other recording configuration will be copied automatically from the Master recorder. Each to connect to AES on its site only. To connect to Master and Central Standby via same network path as used by Avaya components so it will detect failure under same conditions as ESS mode is activated. Set standby.localconfig=true and configure the standby to record the local stations. You may want to have the standby record in Station Bulk mode rather than rely on local TSAPI services. To connect to Master and Standby recorders as their "external controllers". All ports in On Demand mode. Do NOT configure Conferenced mode or any Avaya CT settings. 274 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

275 Standby Recorder Options Master/Standby Topology G In the event of a failure When switchover to Central Standby occurs When remote loses contact with centre Until problem is resolved When problem is resolved Apart from a few seconds delay, calls in progress should be unaffected if the Master fails. If a slave fails, the calls being recorded on it will be reassigned to the remaining slaves within a few seconds. Remote Standby will register phones and start to record as per its local configuration. Standby(s) will continue to record as above. If Central Standby took over, Master will allow the Central Standby to continue recording until you either restart the whole system or instruct it to restore via url /servlet/cscm?cmd=restore. Do this out of hours as there will be a brief interruption to recording as the system switches back. A failed slave, on the other hand, can restored without danger. Note that its ports will be added to the back of the available port list so it may be some time before calls are assigned to it. If remote standby took over due to loss of connection to center, it will continue to record even when connectivity is re-established (since the Avaya system will continue to operate in ESS mode). You must therefore reset the standby at the same time as you restore the Avaya system from ESS mode to centralized control. Do this out of hours as there will be a brief interruption to recording as the system switches back. Issue 5 May,

276 Fault Tolerant Systems Mode of operation This section describes how the standby recorder is configured and how it monitors the health of its primary recorder. It also explains how it takes over when needed and returns to standby mode when appropriate. Standby configuration (automatic) If you do not make alterations to the standby machine's properties file, it automatically copies the license details and settings from its primary. Standby configuration (manual) You only have to configure a standby server manually if you have forced local configuration in the cscm.properties file. A Standby recorder is configured manually in the following way: to match the primary recorder it is to shadow in most respects, for example, system timeouts, maximum recording durations and so on. with the IP addresses of the primary recorder that it is to shadow. Power-On A common case is that all recorders are powered on within a second or two of each other. To avoid destabilizing the system, the standby unit will always wait for a pre-defined period after it starts trying to contact the primary recorder before it assumes that it is dead. This period allows for slight variations in boot time and manual power-on of one unit after another. The time-period is determined by the standby.connecttimeout=xxx property in the properties file. (xxx is in seconds). Standby mode Once configured, the standby unit attempts to establish TCP/IP socket connections to the primary recorder(s) over the one or more IP addresses it has been configured with. If it succeeds, it downloads the configuration details of the primary recorder that are necessary 276 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

277 Mode of operation for it to take over in the event of the primary failing. It continues to refresh these details every minute, thus keeping up to date with configuration changes. Heartbeat messages are exchanged every second. If the standby unit detects a critical error - such as failure to connect to AES - it will shut down its link(s) to the Master so as to avoid any danger of the Master relying on it for fallback when in fact it is unable to function. Failure Detection The standby recorder will attempt to take over from the primary recorder in the following circumstances: The primary requests that the standby unit take over because it has detected a fatal error such as hard disk full, switch connectivity lost completely. On startup, if connection cannot be established with primary after 120 seconds (default - and accurate only to within one minute). All TCP/IP sockets to the master(s) failing and attempts to re-establish them every second are still failing after 30 seconds. Either or both TCP/IP sockets to the Master are still active but the primary does not respond to repeated heartbeat polls for 30 seconds. The three timeout settings above may be overridden by setting the property values standby.connecttimeout, standby.reconnecttimeout and standby.inactivitytimeout respectively in the properties file. Times are specified in seconds. These same parameters are also used by the primary recorder to report corresponding failures of the standby although it will not do anything other than raise an alarm if this occurs. The default values for these timeouts can be overridden on the primary using the same property settings. A primary recorder is not aware of standby recorder(s) until it has established contact with a standby for the first time. Thereafter it will expect to establish contact with that standby recorder always. (This can only be reversed by editing the underlying configuration database). Disk Space Monitoring Once a minute, each recorder will monitor the available disk space on the partitions used for the operating system, call details storage and recording storage. If any of these drops below 500MB a warning will be raised. To override this default; set disk.warnatmb=nnn in the properties file. Issue 5 May,

278 Fault Tolerant Systems Active mode On inferring failure of or being instructed to by the primary, the Standby unit will adopt the appropriate port allocations and attempt to register softphones as per the automatically copied configuration or its local configuration. It will configure the softphones in the same way that the failed recorder had been using them prior to failure (unless you have forced local configuration). Where softphones are marked with specific standby recorder numbers, only those marked for a given standby will be registered. Recordings will be made and uploaded to the eware database or Central Replay Server (if present). Return to Standby mode A standby recorder will return to Standby mode: If the standby unit is shutdown before the failed primary is restored. When the standby unit is rebooted, the primary responds within the timeout mentioned above and the standby recorder remains in standby mode. If the failed unit recovers it may instruct the standby recorder to return to idle immediately - or not until the administrator forces this (see Restoring the Primary on page 279). At this time, the standby unit will truncate all current recordings and un-register all softphones allowing the primary to take over again. Note that this switch-back has the same impact on recordings as the initial switch-over. Hence you should only bring a failed unit back on-line out of hours. Switchover Implications It takes a few seconds to detect most of the failure modes. Although configurable using the properties file, this interval is a compromise between rapid detection of true failure versus risk of false alarms and/or "yo-yoing," a condition where the system goes unstable. The system defaults aim to detect failures within 10 seconds and should not normally be altered. When a failure occurs and the standby recorder becomes active: Recordings in progress are interrupted. The partially completed.wav files for the recordings in progress might be manually recoverable (professional services chargeable). In G.729A mode, up to 1 minute of audio is buffered in memory and hence will not have been appended to the file. In G.711 mode, files are appended to every 30 seconds. 278 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

279 Mode of operation For Station Bulk and Station Executive recording modes: The standby recorder switches in within seconds, registers its softphones and starts recording new calls and calls still in progress from the point at which the softphones are registered and service observe is re-established. For On Demand and Meeting recording modes: Recordings in progress are dropped. The standby recorder switches in within seconds, registers its softphones and awaits new calls arriving on these ports. The worst case loss of recording on a port is therefore determined by the speed with which the switch can re-register the failed softphones and establish service observe again. The more channels on the failed recorder, the longer this process takes. Restoring the Primary The fault condition that caused the standby to become active can be removed in a number of ways, not all of which are ideal. These methods are: Power restored to primary Network connectivity restored to primary Connection to AE Services is restored Primary repaired and reinstated. If there is only one standby recorder, the master recorder will allow it to continue so as not to interrupt recordings unnecessarily. The administrator may force the master to take control again by logging into the administration web pages on the master and then entering the url This will cause the master to take over. It should be performed out of hours as there will be a brief interruption in recording as the system switches over. In cases where multiple standby recorders are, between them, providing backup for a master, this is usually less desirable than having the (single) master regain control. It is also more difficult for the Master recorder to be confident that all recordings are being handled by the standbys. In this case, a recovering Master will retake control automatically - with a brief interruption to service as it does so. The one exception to this is a remote standby configured as part of an ESS. A remote standby will remain recording until it is shutdown. It should be restarted as part of the procedure to unpartition the switch. The master will take control of the stations that were partitioned once the switch is rejoined. Issue 5 May,

280 Fault Tolerant Systems Comparison with hardware switch-over units Traditionally, high availability digital recording has been provided by inserting an "N+1 Switchover Unit" between the recorders and their audio sources. Although this provides a slightly faster switchover, it has several negatives in comparison to the software only approach used here: The cost of the switchover unit makes N+1 systems uneconomical for N=1 or 2 The additional cabling required to and from the switchover unit introduces further cost, failure mechanisms and opportunities for mis-configuration (swapped cables) that might not be noticed until after a failure has occurred. The switchover unit itself introduces a significant single-point of failure into the system. 280 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

281 Standby Recorder Configuration Standby Recorder Configuration Configuration Differences The following Administration pages differ on a Standby recorder: Note: System Settings > License - Check that the license shows as a "Standby" server type. The number of channels is equal to that on the primary recorder that this one is to shadow. Note: If you do not invoke local configuration in the properties file, the standby recorder automatically copies license details from the primary. System Settings > Server - If using a DVD+RW drive for archive, set the path as usual and set the volume label root to a string that is unique to this standby recorder. This will let you distinguish DVDs written on the standby unit from those written on the primaries. - The timeout and purge settings should be kept identical to those on the primary recorder. - An additional setting shows and allows you to enter the IP addresses of the primary recorder that this recorder is to shadow. A primary recorder should have two NIC cards, each with its own IP address. Both addresses should be configured on the standby to enhance the reliability of the network link between them. The default IP port number will be 1414 but you may override this by adding the port number to the IP address, separated by a colon, for example, :1419. If you enter a port number, this must match the configuration of the primary recorder, which also defaults to You can override this default by setting the primary.localport property in the cscm.properties file. Unless you are confident that your DNS server is fault tolerant, you should specify numeric IP addresses to ensure that connectivity does not fail because of address translation problems. - You are strongly advised to create a separate account, using a separate SMTP server from that used by the primary. This way, failure of an SMTP server will be noticed as you continue to receive s from the standby unit but not the daily heartbeats from the primary - or vice versa. It is critical that you ensure that the standby unit is ready to take over at any time. - The IP Station Security Code should be entered the same as for the primary. Important:! Important: Do not use the Advanced settings in the port entry form for the standby recorder. Issue 5 May,

282 Fault Tolerant Systems Security > Users - This page is identical to that on the primary recorder. However, since a central database and replay are nearly always used in systems with standby units, there is little point in entering "replay only" users. Simply use this page to maintain the administrators' account(s). Port Allocations - These pages are provided but their contents are not editable as these are copied from the primary recorder (once contact has been established with it). Status > Alarms and Events - This page is identical to that on the primary recorder though some different alarms will be generated - when the standby establishes contact with the primary and should it ever try to take over from the failed primary. You should use the settings on the System Settings > Server page to ensure that you are advised of these alarms within 10 minutes of one occurring. Status > System Overview - In addition to the normal fields that show on primary recorders, a line will be added to the top table showing the status of the two links to the primary recorder that the standby is configured to shadow. Status > Port States - When the recorder is in standby mode, no ports are visible. When it takes over from the failed primary, this page should be the same as it would have been on the primary had it still been running. Status > Peak Activity - This page is identical to that on the primary recorder but has an added use on the standby recorder. Click on the Restart Peak Activity Count link at the bottom left of this page to zero the counters. When you next view this page, if any of these counts are non-zero you can infer that the standby unit has gone active since you reset the count. This is a quicker check than looking through many pages of alarms and events. 282 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

283 Upgrade Process for Primary/Standby Upgrade Process for Primary/Standby A standby recorder can only act as standby to a primary that is running the same version of recorder software. Please follow these steps when upgrading fault tolerant systems: 1. Take the standby unit OFF the network. 2. Upgrade the primary to the most current version. 3. Confirm correct operation. If any serious problems occur, shut the primary down and reconnect the standby. 4. When confident that the primary is running correctly, upgrade the standby recorder 5. Reconnect the standby to the network. Issue 5 May,

284 Fault Tolerant Systems 284 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

285 Appendix F: AIC Connector This appendix describes the AIC connector. This is a ready-to-use client of the External Control Interface for Avaya Interaction Center. It covers the following topics: Overview on page 286 Installation on page 289 Programmer s Guide on page 290 Issue 5 May,

286 AIC Connector Overview The Verint AIC connector is an AIC server process that allows AIC clients and servers to control the the recorder. It provides the following features to client developers and scriptwriters: Start recording a station Stop recording a station Tag a particular recording with user defined data Enable recording of all calls at a station with automatic tagging Disable recording of that station All these features are made available as simple AIC calls, without the need for complex integration projects. Tagging is a feature of the connector and the recorder that allows clients to associate arbitrary information with each recording. These User-Defined Fields (UDFs), or tags, are stored in the calls details database and can be included in searches and viewed in the search and replay web applications. UDFs are tag-value pairs. (All calls will automatically have the standard set of tag fields.) The connector seamlessly supports: Multiple recorders (if more than one is needed to provide the total number of recording channels) The recorder s primary/standby high-availability mode When there are multiple AIC servers for resilience, the connector can be installed on multiple AIC servers and it will allow recording control of all recorders from either AIC server. The connector itself requires no configuration; it determines its configuration from the recorder s configuration. The recorder is basically configured in the same way that it would be for Station Bulk recording, but with a few minor changes. The AIC connector requires Version 7.7 (or higher) of the recorder. It works in Service Observe mode only. Modes of Operation There are two basic modes of operations: Finer, event driven control Coarser, connector-managed control 286 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

287 Overview In the finer mode, clients of the connector are sent events whenever a call segment starts, and are able to send tagging information via the connector to the recorder for each specific call segment. To operate in this mode a client must register with the connector server for events. It calls the start recording method of the connector. For each call segment that is recorded, the client is sent an event containing the unique recording file serial number. The client can tag each call segment differently if required. The client must stop recording when it is no longer required. In the coarser mode, clients do not need to register for, or handle, events. The connector deals with event handling. In this mode the client enables recording for a particular station and supplies a list of user-defined fields. The connector automatically tags each subsequent call segment for that station with the tags supplied when recording was enabled. The client need take no action, but loses the ability to tag each call segment differently. The client should disable recording when it is no longer required. Principles of Operation The connector is an AIC server process, and runs on the AIC server. It is available for Solaris and Windows. The connector is started and managed by AIC like any server process. The recorders are configured to contact the connector, which listens on a well-known TCP/IP port number. If there is more than one connector, each recorder is configured with all addresses and connects to all of them. The recorders are administered in the usual way for Station Bulk mode, except that they accept external control. The recorder will establish service observe on each target station using one of its recording channels. The target stations can be split across multiple recorders as required. As the recorders connect to the switch and establish Service Observe on the target stations, they announce their ability to record the target stations to the connector. When a client requests recording of a particular station the connector forwards the request to the relevant recorder. At the start of each call segment the recorder sends an event to the connector with the unique serial number of the recording file. The connector sends events to any registered clients and sends tagging information to the recorder as required. High Availability The connector integrates seamlessly with the recorder s primary / standby high-availability arrangement. If the primary recorder fails, it disconnects from the connector, and the secondary connects instead. The connector proceeds to send recording commands to the secondary recorder. Issue 5 May,

288 AIC Connector The recorder can connect to more than one external controller. This allows two connectors to be installed on separate AIC servers. Either connector can be used to route messages to the recorder. 288 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

289 Installation Installation Before the connector can be installed its IDL must be appended to the AIC server s master IDL file, which must then be re-compiled. The resulting pk file must be distributed. The connector s IDL is in the file cscm.idl on the distribution CD. The connector consists of a single executable file that must be copied to the AIC server s directory. It is typically installed in the same directory as the rest of the AIC server executables. Use the AIC admin screens to add an entry for the Verint connector. There are no parameters required. Start the connector on the AIC server(s) where it will run. Configuration Use the ContactStore administration screens to configure the recordable stations. If there is more than one ContactStore, spread the recorded stations evenly across the recorders. Specify the stations or ranges of stations on the Unify page (rather than the Station Bulk page); this allows the connector to stop and start recording. The recorder needs to know that it should continue to request the parties, VDN and UCID information from the switch. Edit the properties file and add the following line: unify.ocpneeded=true Configure the address of the AIC server(s) in the URL field of the Unify page. If there are two AIC servers specify both separated by a semicolon. Check the status of the Unify link(s) on the Status page and correct any alarms. Issue 5 May,

290 AIC Connector Programmer s Guide The interface to the server is described by its IDL, which is shown below. interface CSCM : GeneralS { ORBStatus Enable(in string station, in SeqCouple tags); ORBStatus Disable(in string station); ORBStatus Start(in string station); ORBStatus Stop(in string station); ORBStatus Tag(in string station, in string inum, in SeqCouple tags); ORBStatus Assign(); }; The first two methods (Enable and Disable) are used in the coarser-grained mode of operation; the other four (Start, Stop, Tag and Assign) are used in the finer-grained mode. Note: Note: It is entirely possible to use any names for the tags sent to the recorder, but you should stick to certain conventions. The normal choices for the tag names are spare1 through spare8. If you use these tag names the rest of the Verint infrastructure will make them available throughout the suite of applications. If you use other tag names these will only be accessible through the recorder s own search and replay application. For example, you may wish to tag calls with customer account number and zip code. You should pass a Sequence containing two Couples: spare1=12345, spare2= Even though zip codes can be numbers they should be passed as a string. The Viewer search and replay application allows you to change the names of these fields to more convenient, human-readable names when you design page layouts later. Enable(in string station, in SeqCouple tags) Instructs the connector to enable recording of the specified station, and to tag all subsequent call segments with the tag-value pairs given. If a call is already in progress, recording starts immediately. 290 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

291 Programmer s Guide Disable(in string station) Cancels recording started by an earlier Enable. If a call is in progress, recording is disabled but what was recorded so far is retained. Start(in string station) Instructs the connector to enable recording of the specified station. If a call is in progress, recording starts immediately. Stop(in string station) Cancels recording started by an earlier Start. If a call is in progress, recording is disabled but what was recorded so far is retained. Tag(in string station, in string inum, SeqCouple tags) Instructs the recorder to associate the tags with the recording already in progress (with serial number inum ). Assign() Registers with the connector to receive recording start and stop events. It is these events that contain the inum required for the call-specific tagging. WARNING:! WARNING: Note that the two modes of operation must not be mixed. Clients must either choose the Enable/Disable method, or the Start/Stop/Tag method. Clients may assign a session in either mode in order to receive events. Creating a session is optional with the first method and required with the second. Issue 5 May,

292 AIC Connector 292 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

293 Appendix G: PDS/PCS Connector This appendix describes the PDS/PCS connector. This is a ready-to-use client of the External Control Interface for Avaya Predictive Dialler / Proactive Contact. It covers the following topics: Overview on page 294 Installation on page 295 Issue 5 May,

294 PDS/PCS Connector Overview The Verint PDS/PCS connector is an optional server process that runs on the same server as the recorder. It uses events from the dialler to: Split the otherwise long, single recording of a dialler session into separate recordings, one for each interaction. Tag each of the separate recordings with the dialler Job Name. The PDS/PCS connector requires Version 7.7 (or higher) of the recorder. It works in Station Bulk and Conferenced modes. The connector works with PDS v12, PCS 3.0 and PCS 4.0. When working with PCS 4.0, security can be enabled with a property file entry. Principles of Operation The connector is an optional process that runs on the same server as the recorder.. The recorder is configured to contact the connector, which listens on a well-known TCP/IP port number. A recorder always connects to the connector running on the same server. The recorders are administered in the usual way for Station Bulk or Conferenced mode, except that they accept external control. The connector is automatically informed by the recorder which stations are being recorded. The connector monitors recordings and makes a note of any Agent information. It is able to build a map of which agents are logged on at which station (it does not have access to CTI to determine this in more conventional ways). The connector monitors dialler events for Logon events. If the terminal ID matches either a recorded station or an agent ID that has been seen at a recorded station, the connector notes the recorder port associated with this dialler session. The connector disables recording on that port. When the dialler agent is on a call, recording is re-enabled, and the dialler call information is sent to the recorder. When the dialler agent is between calls recording is again disabled. When the dialler agent logs off, recording control is returned to the recorder. The dialler job name is shown in the Service column of the search and replay screen. 294 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

295 Installation Installation The PDS/PCS connector is part of the recorder installation kit. No additional installation steps are needed. Configuration Use the recorder administration screens to configure the recordable stations. Specify the stations or ranges of stations on the Station Bulk page; this allows the connector to stop and start recording. Add the address of the PDS connector in the URL(s) of Unify/External control port(s) to connect to setting on the System Settings > Server administration page. Specify localhost as the address. The PDS/PCS connector has its own properties file (pdscon.properties). A sample properties file is supplied but must be edited before the connector will work. Edit it and insert site specific data. The following keys are mandatory: PDS.UserName - the PDS client username PDS.Password - the PDS client password PDS.DialerHost - the name of the PDS dialer OAIAddr - the IP address of the recorder s NIC card to be used for communication with the PDS license.key - the license key The following setting is optional: Note: PDS.NSIOR - the Name Service IOR Note: If the IOR is not provided the connector attempts to look up the Name Service using corbaloc (and must be able to resolve the IP address of the dialler by name - make sure it is in DNS) PDS.Secure - true to enable PCS4 security. Issue 5 May,

296 PDS/PCS Connector Choosing the dialler version There are different personality files for the three different versions of PDS/PCS. You must create a symbolic link to enable the correct version. In the /opt/witness/lib directory there are three separate files: pds12_idl.jar, pcs30_idl.jar and pcs40_idl.jar. Create a symbolic link called idl.jar in the same directory to the appropriate jar. For example to select PCS 3.0 as the version enter the command: ln -s pcs30_idl.jar idl.jar Changing the log level Change the log4j.rootlogger key. There are two valid values between the "=" and the "," - DEBUG or INFO. Note that at DEBUG level the log file is likely to fill the disk completely within a few days. Use DEBUG level with caution. Controlling the service You should add the connector as a Linux service. As root, enter chkconfig --add pdscon. If you want the service to start automatically, enter chkconfig --levels 345 pdscon on. As root, you can control the service. service pdscon start service pdscon stop service pdscon restart service pdscon status Automatic restart If the PDS shuts down nightly, consider adding cron entries to stop and start the connector. Shut down the connector after all agents have left and restart the next day before agents begin to log on. 296 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

297 Appendix H: Non-standard Hardware This appendix discusses considerations for non-standard hardware such as blade servers. It covers the following topics: Overview on page 298 Issue 5 May,

298 Non-standard Hardware Overview This manual and the automated installation processes assume a "typical" rack-mounted server as the server hardware. However, ContactStore has been installed on other hardware sucessfully, including blade servers. This appendix covers the considerations for such non-standard hardware. Disks SAN Partitioning Blade servers are normally equipped with locally attached SAN which is ideal for ContactStore. See Storage at Each Recorder on page 43 for further details. The standard installation assumes either one or two physical disks, or one RAID array. In a more complex environment there could be separate arrays for different partitions. For example, a RAID 0 mirror for everything except call storage and a RAID 5 array for call storage. Edit the kickstart script to specify the appropriate disk device for each partition by changing the "on" parameter. See Expert kickstart options on page 84. NICS Blade servers often have redundant network configurations. It is therefore more likely that ContactStore should be installed with just one IP address (eth0) instead of two (eth0 and eth1). DVD Local archive will typically be unavailable since a blade will not have dedicated access to a DVD recorder. Contact Archive is the prefered solution in this environment. 298 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

299 Overview Soundcard The optional local soundcard will likely be unavailable. This is not a problem as a desktop machine can be used to test replay. Kickstart If no floppy disk is available for kickstart installation, use the network-based kickstart instructions in Performing a kickstart install without a floppy on page 83. Issue 5 May,

300 Non-standard Hardware 300 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

301 Appendix I: Advanced Security Settings This appendix discusses some features and prerequisites for advanced security. It includes: Installing Unlimited Strength Encryption on page 302 Installing a Signed SSL Certificate on page 303 Changing Tomcat Port Numbers on page 307 Encrypting Properties File entries on page 308 Configuring Viewer and Archive to use HTTPS on page 309 Configuring Archive for a Key Management Server on page 313 Issue 5 May,

302 Advanced Security Settings Installing Unlimited Strength Encryption To support key sizes larger than 128 bits, you will need to replace the standard Java Jurisdiction Policy Files using the following instructions: 1. Using your browser, go to the following link: 2. Follow the link entitled: Java Cryptography Extension (JCE) Unlimited Strength Jurisdiction Policy Files 5.0 at the foot of the page. 3. Download and unzip the files. 4. Follow the instructions in the readme.txt. (The path for installation is /usr/java/jre1.5.0_13/lib/security). 5. You should note the instruction to back up the existing jars that will be replaced 302 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

303 Installing a Signed SSL Certificate Installing a Signed SSL Certificate If you want to install your own SSL certificate, you must replace the certificate distributed with the application. Your replacement certificate must be specific to your installed server. Selecting a Certificate Authority (CA) If you do not already use a certificate authority, you can use: - FreeSSL requires that the web server has a fully qualified domain name (e.g. contactrecorder.bigcorp.com or contactrecorder.division.bigcorp.com) and needs to be able to send an to an address like [email protected] or [email protected]. The list of addresses can be found on their website, and it includes admin, ssladmin, root, and administrator. - InstantSSL is more flexible and allows intranet addresses (such as WINS names and IP addresses) as well as fully qualified domain names. Backing up the Keystore file In the instructions which follow, replace <installdir> with the location into which you installed Verint ContactStore for Communication Manager. The certificates and keys are stored beneath your installation folder in the file: /opt/witness/keystore/keystore.jks Because this file contains the original, distributed certificate, it is important to make a backup of it. You will delete this file during the remaining steps. Should it be necessary to restore the original certificate, you can copy the backup to the original filename. Creating the new Certificate If you would like to test this implementation, you can practice this procedure with a certificate authority's 30-day trial certificate. Then, to implement real certificates, you can start over from this point. To create a certificate: 1. Create a new certificate with the real URL of the Verint ContactStore for Communication Manager. Issue 5 May,

304 Advanced Security Settings Note: 2. Log onto the server and change directory as follows: cd /opt/witness/keystore 3. Remove the original keystore file rm keystore.jks 4. Run the java keytool utility with /javadirectory/bin/keytool -genkey -keystore keystore.jks -alias tomcat -keyalg RSA 5. Fill in the Keytool prompts with the following: Password: Contact5tor3 Note: You must type this password, exactly as shown. It is case sensitive. a. First & Last Name: enter the FQDN, IP address or intranet name b. Organizational Unit: enter your division c. Organization: enter your company name d. City/Location: enter your location e. State/Province: enter your state f. Country Code: enter the ISO 2 letter code for your country (for example, GB is the code for United Kingdom) 6. Enter yes if the information is correct. 7. Hit enter when prompted for the second password. 8. Restart the Verint ContactStore for Communication Manager service. 9. Access the Administration pages via https. 10. Check that the certificate matches the information entered. 11. Double click the padlock icon. Internet Explorer should warn you that the certificate is unsigned. However, it should no longer display a message that indicates the certificate does not match the web server name. Tip: Tip: If you do get a warning that the certificate does not match, check that the Common Name matches the URL. Double click the padlock, select the details tab, and click the Subject line. This displays the Common Name. 304 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

305 Installing a Signed SSL Certificate Generating a Certificate Signing Request You need a Certificate Signing Request (CSR) as the first step of the signing process. When you have it, paste it into the Certificate Authority's web page. To generate a CSR: 1. Re-run the keytool command /javadirectory/bin/keytool -certreq -keystore keystore.jks -alias tomcat 2. Enter the password - which is Contact5tor3. 3. Copy and paste the output into the CA's web page. (Include the BEGIN and END lines.) 4. Complete the verification process 5. Reply to the verification s and other verification steps until you obtain a signed certificate back from the CA. Importing the CA's certificates Before you can import your certificate reply, you need to import the certificate authority's root certificate and any intermediate certificates between their root and your certificate. To acquire these certificates: 1. Download these certificates from the certificate authority's website. 2. Save the root as rcert.crt and any intermediate as icert.crt. If you have more than one intermediate certificate, give them separate filenames. To import all your certificates: 1. Import the root certificate by running keytool: /javadirectory/bin/keytool -import -keystore keystore.jks -alias root -file rcert.crt 2. Enter the password - which is Contact5tor3. 3. Import the intermediate (if required). /javadirectory/bin/keytool -import -keystore keystore.jks -alias inter -file icert.crt If you have more than one intermediate certificate, import them as inter1, etc. 4. Import your signed certificate. 5. Save the file the CA sent as cert.crt. Issue 5 May,

306 Advanced Security Settings 6. Import with the keytool. /javadirectory/bin/keytool -import -keystore keystore.jks -alias tomcat -file cert.crt 7. Restart the Verint ContactStore for Communication Manager service. 8. Access the administration pages using https. 9. Double click the padlock icon and ensure that Internet Explorer no longer displays a message that the certificate is unsigned. Backing up the keystore file The keystore file now contains: the random private key that is unique to this web server the signed certificate you just paid for These two are linked and cannot be regenerated, so it is important to back up the keystore file. If either one of these components is lost, you must regenerate the certificate and pay again to get it signed. 306 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

307 Changing Tomcat Port Numbers Changing Tomcat Port Numbers You can change the default http and https ports (8080 and 8443 respectively) by editing /opt/witness/tomcat5525/config/server.xml. Locate the two Connector elements on roughly lines 20 and 30 and change 8080 to the chosen plain port number and 8443 to the chosen secure port number. You must also set the following property in the properties file: cscm.localport=nnnn specifying the replacement for Note: Note: If you are using Central Replay Server you must change all the recorders and the Central Replay Server. The port numbers must be consistent across all of these servers for upload to the Central Replay Server to work. If you use Viewer you should alter the URL entered in the Viewer administration screens to your chosen port number. Note: Note: If you change the https port from the default of 8443 and you do not allow http access, you must also set the following line in the properties file so that attempts to use http are redirected to the correct https socket. https.socket=nnnn The server.xml may be overwritten on subsequent upgrades. You should keep a copy of the file after editing it so that this can be compared with any changed version and the appropriate set of merged changes determined after the upgrade. Issue 5 May,

308 Advanced Security Settings Encrypting Properties File entries Verint provides a tool to encrypt your passwords so that they can safely be placed in properties files. You will need this tool if you want to, for example, change the postgresql password. Obtain the WitsBSUserCredentials tool from Verint WAS support and install it following the instructions provided with it. To encrypt a password for use in the properties file: 1. Launch the tool. 2. Select Other for the application type. 3. Enter a dummy username (it is not used) together with the password to be encrypted. 4. Use the encrypted password in the properties file. 308 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

309 Configuring Viewer and Archive to use HTTPS Configuring Viewer and Archive to use HTTPS To allow Viewer and Archive to connect to CSCM using HTTPS, you must first create and install an customised certificate for each recorder in your environment. See Installing a Signed SSL Certificate on page 303. You must obtain certificates with the exact recorder names, as entered in the Capture Platform setting in Viewer. Keep the intermediate and root certificates that you received with the signed certificate and imported into the keystore. You will also need to import these into Viewer and Archive. Importing the public certificates into Viewer and Archive The following procedure describes how to import the public certificates into a windows server. This process must be followed on both the Viewer and the Archive server. This is necessary so that Viewer and Archive can validate the cscm server. 1. Choose Start > Run, type mmc, and then click OK. The Microsoft Management Console (MMC) window opens 2. Click File > Add Remove Snap-in. 3. In the Add/Remove Snap-in window, click Add. 4. In the Add Standalone Snap-in window, click Certificates, and then click Add. 5. Choose Computer Account (to manage the certificate store on the local computer), and then click Next. The Select Computer window opens. 6. Choose Local Computer, and then click Finish. Now that the snap-in is available you should import any intermediate certificate(s). 1. In the Console window under Certificates, right-click Intermediate Certificate Authorities, and then select All Tasks > Import. The Certificate Import Wizard window opens. 2. Click Next. The File to Import window displays. 3. Click Browse 4. Browse for the intermediate certificate file, select the Files of type drop-down list X.509 Certificate (*.cer, *.crt), and select it. 5. Click Next. 6. The Certificate Store dialog box opens. 7. Select Place all certificates in the following store, and then select the Trusted Root Certificate Authorities store click Next. 8. Check the displayed information, and then click Finish. Issue 5 May,

310 Advanced Security Settings Finally import the root certificate. 1. In the Console window under Certificates, right-click Intermediate Certificate Authorities, and then select All Tasks > Import. The Certificate Import Wizard window opens. 2. Click Next. The File to Import window displays. 3. Click Browse 4. Browse for the root.cer file, select the Files of type drop-down list X.509 Certificate (*.cer, *.crt), and select it. 5. Click Next. 6. The Certificate Store dialog box opens. 7. Select Place all certificates in the following store, and then select the Trusted Root Certificate Authorities store click Next. 8. Check the displayed information, and then click Finish. Configure Viewer to use https and a secure port By default Viewer will not use https or the secure port to connect to the cscm recorder, both these features need to be enabled on Viewer. This is done using the Viewer utility EyrBSConfigurationViewer which is located in the following directory on the Viewer server, note by default Viewer uses http and once https is enabled will default to use port C:\Program Files\ComPlus Applications\{A639A F-48E0-B76B-E223C00B5617} Configure Viewer to use https 1. Navigate to the EyrBSConfigurationViewer application and right-click in explorer, select run as 2. The Run As dialog appears, select The following user option and specify the account which Viewer is running under. 3. The Configuration Viewer application is launched. 4. Navigate to Central Configuration > Systems >Main Eyretel System > General 5. In the right-hand pane right-click the SSLOnly value and select Modify 6. Change the value from False to True. 7. Select OK. 310 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

311 Configuring Viewer and Archive to use HTTPS Configure Viewer to use the capture platforms https port 8. Still using the Configuration Viewer, navigate to Central Configuration > Systems > CapturePlatformName > General, where CapturePlatformName is the capture platform name of the cscm recorder as configured in Viewer. 9. In the right-hand pane right-click and select New > Value a new value is created, set the name to SSLHostPort. 10. Now right-click the new value and select Modify 11. Set the value to the https port in use by the cscm recorder, the recorders default https port is Select OK. 13. Now close the Configuration Viewer. 14. Now you should launch Component Services and navigate to Com+ Applications, select Application Portal, right-click and select shutdown. 15. Viewer should now use https when communicating with the cscm recorder. Configure Archive to use https By default Archive will not use https to connect to the cscm recorder, this is done by editing a file located on the Archive server. Configure Archive to use https 1. Open the following file in a text editor C:\Program Files\Verint Systems\conf\SecurityConfig.xml 2. Locate the following xml element in the file <SSLOnly>false</SSLOnly> 3. Change the word false to true and save the file. 4. Now use the Archive Management UI to stop and start the archive service. 5. Archive will now use https to contact to the cscm recorder. As a result of changing this file the archive system will raise two Major alarms indicating that the security configuration file has been tampered and that there is an SSL Configuration Error, both of these can be ignored examples of these are as follows. File Tampered, 04/24/2008 8:44 AM, On Archive781, file "C:\Program Files\Verint Systems\conf\securityconfig.xml" may have been tampered with. Please refer to the System Administration Guide for details on how to correct a tampered file. SSL Configuration Error, 04/24/2008 8:46 AM, SSL is configured for Content Server on host ARCHIVE781, but a required file is missing. The file is C:\Program Files\Verint Issue 5 May,

312 Advanced Security Settings Systems\conf\security\cert.pem. Please deploy the file or remove SSLConfiguration by modifying RecorderGeneral.xml ssllistenport. Note: Note: Archive will use the capture platforms https port as configured using the Viewer Configuration tool as described in the previous section, the error message above is related to a core recorder content server configuration which is not part of the cscm recorder deployment and hence can be ignored. The certificates referred to in the alarm are not necessary for cscm as the public certificates are available via the windows certificate store as described earlier. 312 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

313 Configuring Archive for a Key Management Server Configuring Archive for a Key Management Server Configuring Archive 1. Go to C:\Program Files\Witness Systems\conf\Securityconfig.xml 2. Change the AtRestDataEncryption value to True 3. Go to C:\Program Files\Witness Systems\conf\RSAProvider.cfg 4. Enter the fully qualified domain name of the KMS for kms.address, and change the kms.port to Signing and Delivering the certificate 1. Navigate to the \CAUtility folder on the CA's PC and generate a certificate for the Archive server by typing the following command line at a DOS prompt: gencert <ServerName> <CA's Organization Name> where ServerName is the full server name or just the host name of the server. You will be prompted for: The password of the CA's key file (cakey.pem) An export password to protect the output svr_cert_key.p12 file Successful execution of the above command generates a svr_cert_key.p12 file containing the signed server certificate, its matching private key, and the CA's certificate in PKCS#12 format in the \CAUtility\signed_certs\ServerName folder of the CA's PC. NOTE: The signed server certificate is valid for two years. 2. Deliver the generated file to the C:\Program Files\Witness Systems\conf\Security folder on the Archive server. To ensure security, the export password should not be transported along with the PKCS#12 file. Installing the Certificate To enable SSL, after successful execution of the installation script, you must first execute the following procedure. The procedure converts and stores the necessary certificate files Issue 5 May,

314 Advanced Security Settings for all recorder and JAVA components and must be executed before any other certificate installation procedures. This procedure is required for encryption of data at rest and/or encryption of data in transit. 1. After delivering the svr_cert_key.p12 certificate file to the server, navigate to the C:\Program Files\Witness Systems \conf\security folder and run the following command at a DOS prompt: installcerts export_password where export_password is the entered password when the certificate was signed by the CA. Running the script does the following: Extracts the certificate chain (cert.pem), the server private key (key.pem), and a single server certificate (rsa_app_cert.pem) from the svr_cert_key.p12 file. The rsa_app_cert.pem file needs to be loaded into RSA KMS to identify the key management client (KMC) used by recorder applications on this server if At Rest data encryption with RSA Key Management needs to be enabled. Creates a cacert.pem in the INSTALL_DIR\conf\security\trustedCA directory. Adds the CA certificate (cacert.pem) to the trusted CA certificate keystore (JAVA_HOME\jre\lib\security\cacerts) of the running JRE with Impact360 as the alias. Creates a svr_cert_key_dp.p12 file with the same content as svr_cert_key.p12 except with a new password changeit, which is the default password for the JAVA keystore. Note: Note: Run the above script as often as necessary in case errors occur, such as an incorrect password. 2. Load the file svr_cert_key.p12 into the Windows Personal certificate store of the Computer Account of the server with the export password using the Certificate snap-in of MMC (Start > Run > MMC). See Importing Certificates and Keys on page 65. Make sure that the option Automatically select the certificate store based on the type of certificate is selected. This option adds the server certificate to the Personal certificate store and the CA's certificate to the Trusted Root Certificate Authority certificate store. You may need to refresh the view to see the certificates. 3. Restart Tomcat if applicable 314 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

315 Glossary ACD AE Services AGENT Logon ID ANI Avaya CT BHCA BHCC Communication Manager API (CMAPI) Codec Automatic Call Distribution. This is a feature offered by the Avaya Communication Manager that queues and distributes incoming calls to waiting agents. Calls are queued until an agent is available. If multiple agents are available, calls are distributed on an equitable basis. Avaya's Application Enablement Services are APIs to services such as telecoms, database, and so on. They include DMCC and Avaya CT. An unique identification number assigned to a particular agent. The agent uses this number when logging on. The agent ID is not associated with any particular phoneset Automatic Number Identification (Service). The provision of calling party information (typically, telephone number or billing/account number) to the called party. Avaya Computer Telephony is the CTI link running on the AE Services platform that the recorder uses to observe activity on the Communication Manager and to establish single-step conferences in Conferenced recording mode. It is also used by equality Balance (V5). Avaya CT implements the TSAPI standard and is often referred to as such. Busy Hour Call Attempts. The number of calls that are attempted during the switches busy hour. Typically slightly higher than, but often used interchangeably with BHCC Busy Hour Call Completions. The number of calls that are completed during the switches busy hour. Typically slightly lower than, but often used interchangeably with BHCA Now known as Device, Media and Call Control (DMCC) An abbreviation of COder/DECoder. A device or program that converts signals from one form to another. In this context, between different digital audio compression standards. Issue 5 May,

316 CTI CTI Device, Media and Call Control (DMCC) DID DVD+RW HFS IP IVR JTAPI NAS NIC Computer Telephony Integration - typically an interface through which a computer system can be advised of events occurring within a telephony system and/or control the telephony system. Avaya CT is an example of such a link. A software platform (part of AE Services, previously known as Communciation Manager API or CMAPI) that applications such as the recorder use to create softphones that can participate in calls made on Avaya Communication Manager-based systems. Direct Inward Dialing. An attribute of a trunk. The CO passes the extension number of the called party over a DID trunk to the PBX when offering a call to the PBX. The PBX is then able to automatically route the call to that extension without requiring operator/attendant assistance. In this way, a single trunk can terminate calls for many different extensions (but not simultaneously). Digital Versatile Disk + Read/Write. One of several competing optical storage standards, supported by a wide range of manufacturers. Hierarchical File Storage system. Typically a combination of hard disk and tape drives plus controlling software that automatically migrates little used files to cheaper storage media. Internet Protocol. IP specifies the format of packets and the addressing scheme for internet data. The IP, like the postal system, allows you to address a package and drop it in the system. The packet may traverse multiple networks on the way to its ultimate destination. Interactive Voice Response. A system/facility that plays voice menus to callers, and acts upon user input (typically, DTMF digits from a touch tone phone). It is sometimes called VRU (Voice Response Unit). Java Telephony Application Program Interface. A CTI standard which, in Avaya's case is provided via TSAPI (Avaya CT). Network Attached Storage. A term used for RAID, tape and other mass storage systems which have an integral network connection such as Ethernet or fibre-channel. Network Interface Card. An expansion board that you insert into a computer so the computer can connect to a network. Most NICs are designed for a particular type of network, protocol, and media, although some can serve multiple networks. 316 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide

317 TSAPI Primary Controller QoS RAID SAN Skill Hunt Group Slave Recorder Softphone Standby Controller Tagging Trunk TSAPI A Verint Contact Recorder that is being shadowed by one or more Standby Controllers. The Primary will default to "Active" on startup but will ask a Standby to take over its role if it determines that it cannot record e.g. a disk fills. Quality of Service Redundant Array of Inexpensive Disks. RAID controllers use two or more hard disk drives together for fault tolerance and enhanced performance. RAID disk drives are used frequently on servers. Storage Attached Network. A high-speed network that is typically part of an overall network of computing resources for an enterprise, in which the software knows the characteristics of storage devices and the quantity and value of the data stored in those devices. A telephone number that is used to route calls to agents on the basis of the skills needed to handle the call. Agents are assigned to one or more such skill groups according to their expertise and appropriate calls are routed to them when they are available. A Verint Contact Recorder being controlled by the active Controlling Recorder. i.e. recording what it is told to record. (SOFTware PHONE) In this context, a software-based emulation of an Avaya VoIP phone. Multiple such emulations run on the Verint Contact Recorder and each can participate in a telephone call in order to record it. A Verint Contact Recorder that shadows a Primary Controller and takes over its role ("goes active") should the primary fail or appear to fail (e.g. the standby loses contact with it). Adding details to the database of call recordings so that recordings can later be retrieved by searching through the available data fields. A communications link between a PBX and the public central office (CO), or between PBXs, or between COs. Telephony Services Application Program Interface. A CTI interface standard to which Avaya CT conforms. Issue 5 May,

318 UTC UTC VDN Vector VoIP VPN VRU Universal Time Coordinated. A time scale that couples Greenwich Mean Time, which is based solely on the Earth's inconsistent rotation rate, with highly accurate atomic time. When atomic time and Earth time approach a one second difference, a leap second is calculated into UTC. UTC, like Greenwich Mean Time, is set at 0 degrees longitude on the prime meridian. Vector Directory Number. An extension number used in Avaya's ACD software to connect calls to a vector for processing. The VDN by itself can be dialed to access the vector from any extension connected to the switch. (See also Vector.) A list of steps that processes calls in a user-defined manner. The steps in a vector can send calls to splits, play announcements and music, disconnect calls, give calls a busy signal, or route calls to other destinations. (See also VDN.) Voice over Internet Protocol. A means of transmitting telephone calls over the packet-switched IP network - as distinct from TDM. Virtual Private Network. Private, or restricted, communications networks which use encryption and other security measures to transmit information through a public network such as the Internet and avoid unauthorized use. Voice Response Unit. A device that plays voice menus to a caller and responds to caller instructions entered on a touch tone phone. Also known as Interactive Voice Response (IVR). 318 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide