5nine Cloud Monitor for Hyper-V

Transcription

1 5nine Cloud Monitor for Hyper-V Getting Started Guide Table of Contents System Requirements... 2 Installation... 3 Getting Started... 8 Settings... 9 Authentication nine Cloud Monitor for Hyper-V objects Monitoring settings Alert settings Monitoring Datacenter Level Cluster level Host level Virtual machine level Port level Service level Uninstallation... 44

2 System Requirements Supported Operating Systems: Microsoft Windows Server 2008 R2 SP1 Microsoft Windows Server 2008 R2 Core SP1 Microsoft Hyper-V Server 2008 R2 SP1 Microsoft Windows Server 2012 Microsoft Windows Server 2012 Core Microsoft Hyper-V Server 2012 Prerequisites:.NET Framework 4.0, Microsoft SQL Server.

3 Installation Before installing 5nine Cloud Monitor for Hyper-V on Server Core/Hyper-V operating systems you must perform the following steps: 1) Confirm you have Server 2008 R2 SP1 or later 2) Enable the following features on your Operating System: DISM.exe /online /enable-feature /featurename:servercore-wow64 DISM.exe /online /enable-feature /featurename:netfx2-servercore DISM.exe /online /enable-feature /featurename:netfx3-servercore DISM.exe /online /enable-feature /featurename:netfx2-servercore-wow64 DISM.exe /online /enable-feature /featurename:netfx3-servercore-wow64 DISM.exe /online /enable-feature /featurename:microsoftwindowspowershell DISM.exe /online /enable-feature /featurename:microsoftwindowspowershell-wow64 3) Install.NET 4.0 for Server Core. You can download it at: 4) Use the command net share (or Remote Management) to share folder on Server Core machine, and copy the setup.exe, MainSetup.msi and the license.txt files 5) Execute setup.exe on Server Core machine and follow instructions described below. The Web application will be installed by default to You can then navigate to the console from any machine with a browser:

4 All necessary firewall rules will be added automatically during the installation. Installation package consists of 2 files: 1.) Web application console setup and 2.) agent setup. The agent setup is only needed for a virtual machine inbound and outbound traffic statistics on remote hosts. For other indicators Management console setup is enough, solution is agentless. To setup the Web application the administrator needs to run setup.exe application from the downloaded 5nine Cloud Monitor for Hyper-V archive on the host server or virtual machine that matches the 5nine Cloud Monitor for Hyper-V System Requirements and enter the appropriate license when prompted: Then the 5nine Cloud Monitor for Hyper-V Setup Wizard will be opened, click Next on the dialog window to start installation:

5 Read the 5nine Cloud Monitor for Hyper-V License Agreement and then click Next: Select the folder where 5nine Cloud Monitor for Hyper-V will be installed:

6 You can check the physical space available on your drives and the space required for the installation by pressing the Disk Cost button on the window shown above: Confirm installation and click Next:

7 Enter the credentials of the domain user who has access to Hyper-V machines for the service account: Note: A best practice is to avoid using an existing real user account, but to create a dedicated user with corresponding rights and assign it to the service. Set the SQL Server instance that will be used and connect to it by entering user name and password. There could be either SQL Server Authentication or Windows Authentication used while gaining access to your database. To find out which authentication is used and get the user name and password contact your database administrator. After setup finishes the 5nine Hyper-V Cloud vmonitor service will be started automatically.

8 Getting Started To get 5nine Cloud Monitor for Hyper-V Web application ready to work, the following actions are required: First, the authentication is required prior to do anything further. The authentication as well as master credentials setting processes are described in the Authentication section of Settings chapter. Then, the objects are required to be added for monitoring. This is described in 5nine Cloud Monitor for Hyper-V objects section of the Settings chapter. Monitoring settings adjustment process is described in Monitoring settings section of the Settings chapter. Alert settings adjustment process is described in Alert settings section of the Settings chapter. Monitoring of the objects on the different levels is described in the Monitoring chapter.

9 Settings This chapter describes the necessary settings for preparing 5nine Cloud Monitor for Hyper-V Web application to perfrom as desired. Authentication Upon opening in your browser you will be prompted to enter your authentication credentials: To first logon you should use the credentials under which the service was installed. It s highly recommended to execute the action Master Credentials and set a predefined internal Admin user password. Then for security reasons use only these Master credentials to log on to the console. You can use these Master credentials for direct authentication to REST, OData and PowerShell API in the case of building integration with own systems and when it is not desirable to use/store windows user credentials anywhere. For details about the API integration see the 5nine Hyper-V Cloud vmonitor API.docx file also installed in application folder.

10 5nine Cloud Monitor for Hyper-V objects The monitoring process in 5nine Cloud Monitor for Hyper-V means controlling the certain physical and logical objects by monitoring their state parameters on the corresponding diagrams and checking the alerts. The object types that are monitored by 5nine Cloud Monitor for Hyper-V are described below. Object Types The following objects exist in 5nine Cloud Monitor for Hyper-V: DataCenter. This is the necessary object required to be added prior to any further objects. DataCenter is a formal aggregating entity that exists for the purpose of joining the other objects into the process as its as subsidiaries. Cluster. Cluster object represents multiple servers, two or more, joined into a cluster. Host. Host object represents a single server, which normally hosts virtual machines. Port. Port object represents a standard TCP port, e.g., :3389. Service. Service object represents any OS Windows service, e.g., 59CBTService. Adding objects for monitoring To begin monitoring objects first you must add the DataCenter object. There is a panel menu on the left upper corner: You must specify the name of DataCenter, and optionally Login and Password of windows (domain) user under which performance information will be collected from objects. In most cases you can leave these

11 fields empty then those credentials under which service is running will be used. In the case in which you have several datacenters belonging to different domains then of course you must specify credentials for each of these datacenters. During monitoring these credentials will flow from the root objects to their descendants, however you can always specify some of the descendants have their own credentials. After DataCenter is added, you can select it in the tree view and then the other menus will be un-grayed. On selecting the DataCenter item, there will be un-grayed dropdown menus for adding other types of objects. Click on Add Cluster and enter name, address, login and password in the appropriate fields of the appeared dialog:

12 After a short time (less than 1 minute) you can expand the tree and see its nodes and VMs: Then add host(s):

13 port(s): and service(s):

14 Note: Make sure to enter exactly windows Service name value in the Service Name field of the dialog (do not confuse it with Display Name that is shown in the services list and does not identify the Windows service)! Clicking on the node will display the information summary for the selected object at the main Summary tab. The current status of the selected object is also displayed on the Summary tab. You can also Enable or Disable monitoring of the selected object by simply clicking the Enable/Disable Monitoring button which appears on the right-upper spot of the window depending on the current status: The view will differ depending on which object is selected. The sections of the Monitoring chapter below will describe the monitoring tabs view on all levels.

15 Other panel menu commands If you want to edit the selected object, click Edit Object button on the left panel: Then enter new parameters in the corresponding dialog that will appear depending on what object is selected: for datacenter: for cluster:

16 for host: for port: for service:

17 If you want to set or change credentials for the selected object, click the Set Credentials button on the left panel: Then enter your login and password in the appeared dialog as required and click OK. If you want to remove the selected object, click the Remove Object button on the left panel: Then confirm the removal of the object in the dialog shown: If you want to refresh the list of monitored objects, click the Refresh List button on the left panel: If you want to refresh the information of the selected object, click the Refresh Information button on the right panel:

18 If you finished working with the application you can close the session, by clicking the Log Out button in right-upper corner of the 5nine Hyper-V Cloud vmonitor main window: By default if there is no any activity for a period of 20 minutes thesession will be closed automatically.

19 Monitoring settings You can change the global monitoring settings, set up SMTP alert actions and mail settings by clicking the Monitoring Settings button in right-upper corner of the 5nine Cloud Monitor for Hyper-V main window:

20 Set the monitoring parameters as follows: Clear the Enable Guest Monitoring checkbox if you wish to exclude virtual machines from monitoring. All the other added objects will still be monitored. Set the monitoring frequency: the values are chosen from the Poll frequency list box (1,2,3,4,5,10,15 minutes): Enter credentials in the fields Login and Password, if needed. It is not necessarily to do this in the case in which the service is installed already has permissions to collect information on monitored objects. Set alert actions for warnings and errors. There are two similar list boxes for each alert type with the following choices: None do not document alerts. Add Event to Event Log add events to log (will be displayed on Alerts tab of main window). Send Notification send to the chosen recipients (settings described below). Add Event and Send Notification perform both of the actions above. Set up mail parameters for notifications: Specify SMTP server address (e.g., mail.crossinform.ru or ). Enter the port. Tick Requires Authentication box if authentication is necessary. Tick SSL box if security socket layer usage required when sending notification s. Enter credentials user name and password if SMTP server requires authentication (Requires Authentication box was ticked as described above). Specify address from which notification s will be sent. Enter addresses of recipients, separated by a semicolon (;).

21 Alert settings You can browse alerts on a per object basis by clicking the corresponding objects in the tree, or you can select datacenter and see all alerts for the datacenter s descendant objects: You can dismiss some of alerts or all at once by clicking corresponding clear alerts buttons: All of the actions with alerts are similar no matter which object is selected.

22 If you want to overview the current alert settings click Alert Settings button in right-upper corner of the 5nine Cloud Monitor for Hyper-V main window: If you want to change the setting of the alert then just click the pencil button icon on the left of the alert list. The following dialog will appear:

23 Set the alert parameters as follows and (change predefined values if necessary): Enter the alert name. There are predefined names for all the factors, though you can alter them at your own wish. The new alert name will be shown then in all other places. Enter the new description. Choose the new category. Values are chosen from the Category list box: None neither of the parameters will be monitored. Health parameter health will be monitored. Processor processor state/usage will be monitored. Memory memory utilization/availability will be monitored. Network network utilization/performance will be monitored. Storage disk utilization/performance will be monitored. Chose the scope for monitoring. Values are chosen from the Scope list box: None neither level will be read. Host parameters will be read on the host level. Guest parameters will be read on the VM level. Cluster parameters will be read on the cluster level. Node parameters will be read on the node level. Port parameters will be read on the port level. Service parameters will be read on the service level. DataCenter parameters will be read on Data Center level. Global parameters will be read on the global level. Choose the alert condition from the Condition list box: Equals the reading must be equal to the set threshold value for the alert. Above the reading must exceed the set threshold value for the alert. Below the reading must be below the set threshold value for the alert.

24 Choose the unit of measure from the UOM list box: Percent for the percent. KB for the kilobytes. MB for the megabytes. GB for the gigabytes. KBPerSec for the kilobytes per second. MBPerSec for the megabytes per second. GBPerSec for the gigabytes per second. Count for counting the number of necessary units/events etc. Trigger Rule. It is advised not to alter this predefined parameter. Enter warning threshold for the alert. This figure refers to the numerical value of the alert parameters entered above for the warning alert to happen. Enter error threshold for the alert. This figure refers to the numerical value of the alert parameters entered above for the error alert to happen. Enter the condition length in seconds that must be reached for the alert. Enter the report range. Enter the report frequency in seconds. Clear Enabled check box if you no longer wish for the current alert to happen. Click OK in the Modify Alert Definition dialog when you are finished. The table below contains the description of predefined failures and matching parameters: Parameter Failure Description High Logical Processor Usage Host processor overutilization A condition when the percentage of time spent by the processor in guest and hypervisor code is greater than specified threshold High Memory Pressure Host memory low performance A condition when the rate at which pages are read from or written to disk is greater than specified threshold. Low Network Performance Network throughput insufficiency A condition when the network interface output queue length is greater than specified threshold. High Disk I/O Utilization Host disk input/output overload A condition when condition when the disk I/O utilization is greater than specified threshold.

25 Parameter Failure Description High Virtual Processor Usage Low Disk Performance Host Connection Failure High VP to LP Ratio Low Available Memory VM Health Critical High Network Utilization High Virtual Network Utilization High Daily Inbound Network Traffic High Guest Processor Usage High Guest Processor Frequency Low Dynamic Memory Buffer High Assigned Memory Usage Low Free Space Overutilization of host processor resources assigned for virtual machines Host disk performance insufficiency Impossibility to set the connection with the host in a timely manner Over usage of host processor resources in terms of its share assigned to virtual machines Critically low amount of available memory on the host Greater amount of VMs in critical health Network resources comparative overutilization Virtual network traffic is over limit Virtual daily inbound network traffic is over limit Over usage of a host processor resources assigned for virtual machine in percentage Frequency of a host processor when serving virtual machine is over limit Available dynamic memory buffer for virtual machine is below limit Assigned virtual machine memory overutilization Disk free space assigned for virtual machine is below limit A condition when the percentage of time spent by the virtual processor in guest and hypervisor code is greater than specified threshold. A condition when the storage output queue length is greater than specified threshold. A condition when the host connection cannot be established during the specified threshold timeout. A condition when the virtual to logical processors ratio is greater than specified threshold. A condition when the percentage of available memory is less than specified threshold. A condition when the number of virtual machines that with critical health is greater than specified threshold. A condition when the percentage of network utilization is greater than specified threshold. A condition when the virtual network adapter utilization is greater than specified threshold. A condition when the inbound network traffic from the beginning of the current day is greater than specified threshold. A condition when guest processor load is greater than specified threshold. A condition when guest processor frequency is greater than specified threshold. A condition when available memory buffer is less than specified threshold. A condition when the assigned memory usage is greater than specified threshold. A condition when the free space is lower than specified threshold.

26 Parameter Failure Description High Daily Outbound Network Traffic High Virtual Disk Utilization High Memory Usage Cluster Connection Failure Port Connection Failure Service Failure Virtual daily outbound network traffic is over limit Virtual disk resources overload Assigned virtual memory over usage in percent Impossible to set the connection with the cluster in a timely manner Impossible to set the connection through the port in a timely manner Service unavailability or inactive state (not running) during specified time period A condition when the outbound network traffic from the beginning of the current day is greater than specified threshold. A condition when the virtual disk utilization is greater than specified threshold. A condition when the memory usage is greater than specified threshold. A condition when the cluster connection cannot be established during the specified threshold timeout. A condition when the port connection cannot be established during the specified threshold timeout. A condition when the service cannot be found or state is not running during the specified threshold timeout.

27 Monitoring The chapter describes monitoring the interface view for different objects by levels. Datacenter Level When the selected object is Datacenter there will be the following tab information view: Summary

28 The information shown here differs from that on the single host as datacenter is supposed used to join multiple objects. The total figures are displayed here with the top 3 resource consumers shown for each performance parameter. Alerts Alerts tab displays the list of alerts for the selected datacenter:

29 Cluster level When the selected object is a Cluster there will be the following tab information view: Summary The information shown on the Summary tab for a cluster looks to the same as that for a datacenter. There are total figures are displayed here with top 3 resource consumers shown for each performance parameter.

30 Alerts Alerts tab displays the list of alerts for the selected cluster:

31 Host level When the selected object is the host there will be the following tab information view: Summary

32 CPU CPU tab displays recent CPU information for the selected host: The upper diagram shows the host logical processor usage in percent with the predefined poll frequency either by default value or set by user (will be described below). The lower diagram shows the CPU utilization percentage by virtual machines running on the current host.

33 Memory Memory tab displays recent memory information of the selected host: The diagram in the picture shows the total percentage of memory utilization on the current host. Network Network tab displays recent network utilization data of the selected host: The diagram in the picture shows the total percentage of network activity on the current host.

34 Disk Disk tab displays recent disk utilization data for the selected host: The diagram in the picture shows the total disk usage information on the current host in real figures (bps/kbps/mbps). Alerts Alerts tab displays the list of alerts for the selected host:

35 Virtual machine level The Virtual machine must be on and running so that all the tabs for it are displayed. When the selected object is a virtual machine there will be the following tab information view: Summary

36 CPU CPU tab displays recent CPU information for the selected virtual machine: The diagram shows the host CPU utilization percentage share of current virtual machine. Memory Memory tab displays recent memory information for the selected virtual machine: The upper diagram shows the dynamic memory buffer availability in percent for the current virtual machine. The lower diagram shows the guest memory utilization percentage for the current virtual

37 machine. Network Network tab displays recent network utilization data for the selected virtual machine. There are 3 diagrams in the Network tab: Virtual Network Usage

38 Outbound Traffic Outbound traffic diagram displays the top IP statistics for outbound traffic, i.e. top IP-address traffic contributors are shown in descending order for a selected VM. Percentage shows the frequency IP-address is the top traffic contributor within the period shown in the diagram. Inbound Traffic Outbound traffic diagram displays similar information but for inbound traffic for a selected VM.

39 Note: The last two diagrams are shown only if the 5nine Monitor capturing extension is installed and switched on. Disk Disk tab displays recent disk utilization data for the selected virtual machine: The upper diagram shows the virtual disk usage information on the current virtual machine in real figures (bps/kbps/mbps). The lower diagram shows the free space on the virtual disk using the total memory share assigned for current virtual machine from the host as the total available.

40 Win2012 Resource Metering Win2012 Resource Metering tab shows the new parameters specifically for the virtual machines of the host with OS Windows Server 2012 installed. It currently shows the host CPU and memory usage parameters for the current virtual machine as it appears in the display picture: - Average CPU Usage (in MHz); - Average memory use (in Mb); - Minimum memory use (in Mb).

41 Alerts Alerts tab displays the list of alerts for the selected virtual machine:

42 Port level When the selected object is a port the following tab information view will display: Summary Alerts Alerts tab displays the list of alerts for the selected port:

43 Service level When the selected object is a service the following tab information view will display: Summary Alerts Alerts tab displays the list of alerts for the selected service:

44 Uninstallation To uninstall 5nine Cloud Monitor for Hyper-V from your server, go to Control Panel Programs Programs and Features Uninstall a Program. Then choose 5nine Hyper-V Cloud vmonitor on the program list and click Uninstall to run 5nine Cloud Monitor for Hyper-V uninstallation process, and then follow the system hints.