Oracle Grid Engine. Installation and Upgrade Guide Release 6.2 Update 7 E

Transcription

1 Oracle Grid Engine Installation and Upgrade Guide Release 6.2 Update 7 E February 2012

2 Oracle Grid Engine Installation and Upgrade Guide, Release 6.2 Update 7 E Copyright 2000, 2012, Oracle and/or its affiliates. All rights reserved. Primary Author: Uma Shankar Contributing Author: Contributor: Andy Schwierskott This software and related documentation are provided under a license agreement containing restrictions on use and disclosure and are protected by intellectual property laws. Except as expressly permitted in your license agreement or allowed by law, you may not use, copy, reproduce, translate, broadcast, modify, license, transmit, distribute, exhibit, perform, publish, or display any part, in any form, or by any means. Reverse engineering, disassembly, or decompilation of this software, unless required by law for interoperability, is prohibited. The information contained herein is subject to change without notice and is not warranted to be error-free. If you find any errors, please report them to us in writing. If this is software or related documentation that is delivered to the U.S. Government or anyone licensing it on behalf of the U.S. Government, the following notice is applicable: U.S. GOVERNMENT RIGHTS Programs, software, databases, and related documentation and technical data delivered to U.S. Government customers are "commercial computer software" or "commercial technical data" pursuant to the applicable Federal Acquisition Regulation and agency-specific supplemental regulations. As such, the use, duplication, disclosure, modification, and adaptation shall be subject to the restrictions and license terms set forth in the applicable Government contract, and, to the extent applicable by the terms of the Government contract, the additional rights set forth in FAR , Commercial Computer Software License (December 2007). Oracle America, Inc., 500 Oracle Parkway, Redwood City, CA This software or hardware is developed for general use in a variety of information management applications. It is not developed or intended for use in any inherently dangerous applications, including applications that may create a risk of personal injury. If you use this software or hardware in dangerous applications, then you shall be responsible to take all appropriate fail-safe, backup, redundancy, and other measures to ensure its safe use. Oracle Corporation and its affiliates disclaim any liability for any damages caused by use of this software or hardware in dangerous applications. Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners. Intel and Intel Xeon are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license and are trademarks or registered trademarks of SPARC International, Inc. AMD, Opteron, the AMD logo, and the AMD Opteron logo are trademarks or registered trademarks of Advanced Micro Devices. UNIX is a registered trademark of The Open Group. This software or hardware and documentation may provide access to or information on content, products, and services from third parties. Oracle Corporation and its affiliates are not responsible for and expressly disclaim all warranties of any kind with respect to third-party content, products, and services. Oracle Corporation and its affiliates will not be responsible for any loss, costs, or damages incurred due to your access to or use of third-party content, products, or services.

3 Contents Preface... ix Audience... Documentation Accessibility... Related Documents... Conventions... ix ix ix ix 1 Planning the Installation System Requirements Disk Space Requirements Supported Operating Platforms Planning Checklist Cluster Design Cells Cluster Name Queue Structure Host System Requirements Master Host Shadow Master Hosts Execution Hosts Administration Hosts Submit Hosts User Account Considerations User Names Installation Accounts File Access Permissions Network Services Installation Methods Directory Organization Spool Directories under the Root Directory Choosing Between Classic Spooling and Database Spooling $SGE_ROOT Directory Spooling Options Database Server and Spooling Host Scheduler Profiles Getting the Software iii

4 Electronic Download CD-ROM Distribution Installing Grid Engine Loading the Distribution Files on a Workstation How to Load the Distribution Files on a Workstation pkgadd Method tar Method Installing the Software With the GUI Installer Requirements Express Installation Using the Express Installation Mode Custom Installation Using the Custom Installation Mode How to Configure Password-less Access for the root User Configuring Password-less ssh Access for the root User Configuring Password-less rsh Access for the root User Understanding Host and Installation States Host Resolving Host Validating Installation States Tweaking start_gui_installer Description of start_gui_installer Options Using start_gui_installer Options installing as a Different connect_user Installing Single Windows Execution Host Installing Multiple Windows Execution Hosts Troubleshooting the GUI Installer FAQs Known issues and workarounds Installing the Software From the Command Line Installation Overview Performing an Installation How to Install the Master Host Installing the Master Host Example Master Host Installation How to Install Shadow Master Host Starting a Shadow Master Host Manually Configuring Shadow Master Host Environment Variables Example Shadow Master Host Installation How to Install Execution Hosts Example Execution Host Installation How to Register Administration Hosts How to Register Submit Hosts How to Install the Berkeley DB Spooling Server Installing the Increased Security Features Why Install the Increased Security Features? iv

5 Additional Setup Required How to Install a CSP-Secured System How to Generate Certificates and Private Keys for Users How to Renew Certificates How to Check Certificates Displaying a Certificate Check Issuer Check Subject Show of Certificate Show Validity Show Fingerprint Verifying the Installation How to Verify That the Daemon is Running on the Master Host How to Verify That the Daemons Are Running on the Execution Hosts How to Run Simple Commands How to Submit Test Jobs Automating the Installation Process Automatic Installation Special Considerations Using the inst_sge Utility and a Configuration Template How to Automate Installation With Increased Security (CSP) How to Automate Other Installations Through a Configuration File How to Automate the Master Host Installation Automating Other Installations Through a Configuration File Automatic Uninstallation How to Uninstall Execution Hosts Automatically How to Uninstall the Master Host Automatically How to Uninstall the Shadow Master Host How to Start the Automatic Backup Troubleshooting Automatic Installation and Uninstallation Installing SMF Services Why Install SMF Services? Additional Setup Required How Do SMF Services Compare to the Normal Services? qmaster Daemon shadowd Daemon execd Daemon Berkeley RPC Server dbwriter Software Installing a JMX-Enabled System Additional Setup Required How to Install a JMX Agent-Enabled System How to Generate Certificates, Private Keys and Keystores for Users How to Check Certificates, Private Keys and Keystores for Users JMX Configuration Files jaas.config java.policy v

6 vi management.properties jmx.access jmx.password logging.properties Testing and Troubleshooting Removing the Software How to Remove the Software Interactively How to Remove the Software Using the inst_sge Utility and a Configuration Template Additional Software for the Microsoft Operating System Additional Software Microsoft Services for UNIX Unsupported Grid Engine Functionality Configuring User Name Mapping How to Install Microsoft Services for Unix System Requirements Services for UNIX Installation Post SFU Installation Tasks Troubleshooting SFU Microsoft Subsystem for UNIX-based Applications Unsupported Grid Engine Functionality How to Install a Microsoft Subsystem for UNIX-based Applications System Requirements Installing Subsystem for UNIX-based Applications Post Installation Tasks Troubleshooting Microsoft Subsystem for UNIX-based Applications Changing Default Behavior to Case Sensitivity Disabling DEP How to Disable DEP for Windows XP Professional, Windows Server 2000 and Window Server How to Disable DEP for Windows Vista (Enterprise and Ultimate) and Windows Server Enabling suid Behavior for Interix Programs User Management on Windows Hosts Managing Users on Windows Hosts Windows User Example UNIX User Management Using Grid Engine in a Microsoft Windows Environment Registering Windows User Passwords Using the sgepasswd Command Adding Windows Hosts to Existing Grid Engine Systems How to Add Windows Hosts Later Other Installation Issues How to Verify and Install Linux Motif Libraries How to Install the Software on a System with IPMP What Is IP Multipathing? Issues Between IPMP and Grid Engine Installing the Grid Engine Master Node With IPMP Ignoring the Error Messages

7 Temporarily Disabling IPMP Installing a Grid Engine on an Execution Host With IPMP Enabling Administrative and Submit Hosts With IPMP Upgrading Grid Engine About Upgrading the Software Before You Upgrade Constraints How to Back Up the Configuration of the Old Cluster What the Backup Contains How to Back Up the Cluster How to Install the 6.2 Software Using the Cloned Configuration Method Additional Constraints for the New 6.2 Installation with Cloned Configuration Example Upgrade for Cloned Cluster Configuration How to Upgrade the Original Cluster to the 6.2 Software (Real Upgrade) How to Upgrade from 5.3 to A Configuration File Templates Configuration File Template... A-1 vii

8 viii

9 Preface Oracle Grid Engine Installation and Upgrade Guide describes how to install Grid Engine and how to upgrade the Grid Engine from a previous version. Audience This document is intended for system administratrs to perform installation or upgrading Oracle Grid Engine. Documentation Accessibility For information about Oracle's commitment to accessibility, visit the Oracle Accessibility Program website at Access to Oracle Support Oracle customers have access to electronic support through My Oracle Support. For information, visit or visit if you are hearing impaired. Related Documents Conventions For more information, see the following documents in the Oracle Grid Engine Release 6.2 documentation set: Oracle Grid Engine Release Notes Oracle Grid Engine User Guide Oracle Grid Engine Administration Guide The following text conventions are used in this document: Convention italic monospace Meaning Italic type indicates book titles, emphasis, or placeholder variables for which you supply particular values. Monospace type indicates commands within a paragraph, URLs, code in examples, text that appears on the screen, or text that you enter. ix

10 x

11 1 1Planning the Installation Before you install the Grid Engine software, you must plan how to achieve the results that fit your environment. This section helps you make the decisions that affect the rest of the procedure. System Requirements Disk Space Requirements To verify that the systems on which you intend to install Grid Engine conform to required hardware and software specifications, review the system requirements listed below. The Grid Engine software directory tree has the following fixed disk space requirements: 50 Mbytes for the installation files without any binaries Between 60 and 100 Mbytes for each set of binaries The ideal disk space for Grid Engine system spool directories is as follows: Mbytes for the master host spool directories Mbytes for the Berkeley DB spool directories The spool directories of the master host and of the execution hosts are configurable and need not reside under the default location, sge-root. Note: You must satisfy several Windows platform-specific prerequisites before you can install Grid Engine on hosts that are running the Windows operating system. You might need to install additional software on your computer which might require additional disk space. See Microsoft Services for UNIX and Microsoft Subsystem for UNIX-based Applications. Planning the Installation 1-1

12 Planning Checklist Supported Operating Platforms The Grid Engine 6.2 software supports the following operating systems and platforms: Master Host Solaris 11, 10, 9, and 8 Operating Systems (SPARC Platform Edition) Solaris 9 Operating System (x86 Platform Edition) Solaris 11 and 10 Operating Systems (x64 Platform Edition) Linux x86, kernel 2.4 and higher, glibc >= Linux x64, kernel 2.4 and higher, glibc >= Compute Host Solaris 11, 10, 9, and 8 Operating Systems (SPARC Platform Edition) Solaris 9 Operating System (x86 Platform Edition) Solaris 11 and 10 Operating Systems (x64 Platform Edition) Linux x86, kernel 2.4 and higher, glibc >= Linux x64, kernel 2.4 and higher, glibc >= Linux IA64, kernel 2.4, 2.6, glibc >= Apple Mac OS X 10.4 (Tiger), PPC platform Apple Mac OS X 10.4 (Tiger), x86 platform Apple Mac OS X 10.5 (Leopard), x86 platform Hewlett Packard HP-UX or higher, 32 bit Hewlett Packard HP-UX or higher, 64 bit (including HP-UX on IA64) IBM AIX 5.1, 5.3, 6.1 Microsoft Windows: Server 2003 XP Professional with at least Service Pack Server with at least Service Pack Professional with at least Service Pack 3 Server 2003 Release 2 Server 2008 Vista Enterprise Vista Ultimate Planning Checklist Before you install the Grid Engine software, you must plan how to achieve the results that fit your environment. This section helps you make the decisions that affect the rest of the procedure. Write down your installation plan in a table similar to the following example. 1-2 Oracle Grid Engine Installation and Upgrade Guide

13 Cluster Design Table 1 1 Parameter $SGE_ROOT directory Cell name Planning Checklist $SGE_CLUSTER_NAME Administrative user sge_qmaster port number (6444 is recommended) sge_execd port number (6445 is recommended) Master host Shadow master hosts Execution hosts Spooling for each execution host (global or local) Windows execution hosts (yes or no) Administration hosts Submit hosts Group ID range for jobs Spooling mechanism (Berkeley DB or Classic spooling) Berkeley DB server host (the master or another host) Berkeley DB spooling directory on the database server Scheduler tuning profile (Normal, High, Max) Installation method (interactive, secure, automated, or upgrade) Value Cluster Design If you are going to install Grid Engine 6.2 on Microsoft Windows Server 2003, Windows XP Professional with at least Service Pack 1, Windows 2000 Server with at least Service Pack 3, or Windows 2000 Professional with at least Service Pack 3, acquire and install Microsoft Services For UNIX. See Microsoft Services for UNIX for more information. If you are going to install Grid Engine 6.2 on Microsoft Windows Server 2003 Release 2, Windows Server 2008, Windows Vista Enterprise or Windows Vista Ultimate, acquire and install Microsoft Subsystem for UNIX-based Applications. See Microsoft Subsystem for UNIX-based Applications for more information. If you are going to install Grid Engine 6.2 on a Windows system, create the required Certificate Security Protocol (CSP) certificates before installing Grid Engine. See How to Install a CSP-Secured System for information about CSP certificates. Check Other Installation Issues for applicability. Planning the Installation 1-3

14 Queue Structure Cells You can set up the Grid Engine system as a single cluster or as a collection of loosely coupled clusters called cells. The $SGE_CELL environment variable indicates the cluster being referenced. When the Grid Engine system is installed as a single cluster, $SGE_ CELL is not set, and the value default is assumed for the cell value. Cluster Name The $SGE_CLUSTER_NAME environment variable supports unique naming of the cluster. Unlike the $SGE_CELL variable, there are restrictions on $SGE_CLUSTER_NAME. If you decide to use Grid Engine SMF services on Solaris 10 or later hosts, you must select a new $SGE_CLUSTER_NAME. This name becomes part of the name of the Grid Engine SMF services. The $SGE_CLUSTER_NAME is also used to distinguish multiple rc files for different clusters. Note: If your $SGE_CELL name already reflects the desired cluster name and also satisfies $SGE_CLUSTER_NAME restrictions, set the cluster name to the $SGE_CELL value. Otherwise, the proposed default value is psge_qmaster_port, which uniquely identifies the running cluster by the port on which its qmaster daemon is running. See Installing SMF Services for more information. Queue Structure The installation procedure creates a default cluster queue structure, which is suitable for getting acquainted with the system. The default queue can be removed after installation. Note: No matter what directory is used for the installation of the software, the administrator can change most settings that were created by the installation procedure. This change can be made while the system is running. Consider the following when determining a queue structure: Whether you need cluster queues for sequential, interactive, parallel, and other job types Which queue instances to put on which execution hosts How many job slots are needed in each queue For more detailed information on administering cluster queues, see Oracle Grid Engine Administration Guide. Host System Requirements Master Host The master host controls the Grid Engine system. This host runs the master daemon sge_qmaster. The master host must comply with the following requirements: 1-4 Oracle Grid Engine Installation and Upgrade Guide

15 Host System Requirements The host must be a stable platform. The host must not be excessively busy with other processing. At least 60 to 120 Mbytes of unused main memory must be available to run the Grid Engine system daemons. For very large clusters that include many hundreds or thousands of hosts and tens of thousands of jobs in the system at any time, 1 GByte or more of unused main memory might be required and 2 CPUs might be beneficial. The master host must be installed before shadow master execution, administration, or submit hosts. (Optional) The Grid Engine software directory, $SGE_ROOT, should be installed locally to cut down on network traffic. Note: Windows hosts cannot act as master hosts. For more information, see How to Install the Master Host. Shadow Master Hosts These hosts back up the functionality of sge_qmaster in case the master host or the master daemon fails. To be a shadow master host, a machine must have the following characteristics: It must run sge_shadowd. It must share sge_qmaster status, job information, and queue configuration information that is logged to disk. In particular, the shadow master hosts need read/write root or administration user access to the sge_qmaster spool directory and to the $SGE_ROOT/$SGE_CELL/common directory. The $SGE_ROOT/$SGE_CELL/common/shadow_masters file must contain a line defining the host as a shadow master host. Note: If no cell name is specified during installation, the value of $SGE_CELL is default. The shadow master host facility is activated for a host as soon as these conditions are met. You do not need to restart the Grid Engine system daemons to make a host into a shadow master host. Note: Windows hosts cannot act as shadow master hosts. For more information, see How to Install Shadow Master Host. Execution Hosts Execution hosts run the jobs that users submit to the Grid Engine system. An execution host must first be set up as an administration host. You run an installation script on each execution host. For more information, see How to Install Execution Hosts. Planning the Installation 1-5

16 User Account Considerations Administration Hosts Operators and managers of the Grid Engine system use administration hosts to perform administrative tasks such as reconfiguring queues or adding Grid Engine users. The master host installation script automatically makes the master host an administration host. During the master host installation process, you can add other administration hosts. You can also manually add administration hosts on the master host at any time after installation. Submit Hosts Jobs can be submitted and controlled from submit hosts. The master host installation script automatically makes the master host a submit host. User Account Considerations User Names For the Grid Engine system to verify that users submitting jobs have permission to submit them on the desired execution hosts, users' names must be identical on the submit and execution hosts. You might therefore have to change user names on some machines, because Grid Engine user names map directly to system user accounts. Note: User names on the master host are not relevant for permission checking. These user names do not have to match or even exist. Installation Accounts You can install the Grid Engine software either as the root user or as an unprivileged user, for example, your own user account. However, if you install the software when you are logged in as an unprivileged user, the installation allows only that user to run Grid Engine jobs. Access is denied to all other accounts. Installing the software when you are logged in as root resolves this restriction. However, root permission is required for the complete installation procedure. Also, if you install as an unprivileged user, you are not allowed to use the qrsh, qtcsh, or qmake commands, nor can you run tightly integrated parallel jobs. To use SMF on Solaris 10 or later hosts and run the Grid Engine software as an unprivileged user, perform the following additional steps as root user (or user with appropriate permissions): 1. Create the new role sgeadmin for the local user : roleadd -c "Grid Engine SMF Administrator" -g <group> -d <home_dir> -u <UID> -s <profile_shell> -P "solaris.smf.manage.sge" "sgeadmin" 2. Assign the just-created role sgeadmin to the user: usermod -R "sgeadmin" <login> For a distributed name service, such as NIS, NIS+, or LDAP, create the new role sgeadmin and assign it to the user: /usr/sadm/bin/smrole add -D <domain_name> - -n "sgeadmin" -a "normal_user" -d 1-6 Oracle Grid Engine Installation and Upgrade Guide

17 Installation Methods <home_dir> -c "Grid Engine SMF Administrator" -p "solaris.smf.manage.sge" File Access Permissions If you install the software logged in as root, you might have a problem configuring root read/write access for all hosts on a shared file system. Therefore, you might have problems putting the $SGE_ROOT files onto a network-wide file system. You can force Grid Engine software to run all Grid Engine system components through a non-root administrative user account, for example sgeadmin. With this setup, this particular user needs only read/write access to the shared $SGE_ROOT file system. The installation procedure asks whether files should be created and owned by an administrative user account. If you answer "Yes" and provide a valid user name, files are created by this user. Otherwise, the user name under which you run the installation procedure is used. Create an administrative user, and answer "Yes" to this question. Make sure in all cases that the account used for file handling on all hosts has read/write access to the $SGE_ROOT directory. Also, the installation procedure assumes that the host from which you access the Grid Engine software distribution media can write to the $SGE_ROOT directory. Note: The name of the root user on Windows hosts depends on the system language of the Windows operating system. You can even change the name of the root user. The default name for many languages is the name Administrator. If your Windows host is a member of a Windows domain, only the local Administrator is the root user. Neither the members of the Administrators group, nor the domain Administrator, nor a member of the Domain Admins group are the root user. See User Management on Windows Hosts for more information about users on Windows hosts. Network Services Determine whether your site's network services are defined in an NIS database or in an /etc/services file that is local to each workstation. If your site uses NIS, determine the host name of your NIS server so that you can add entries to the NIS services map. The Grid Engine system services are sge_execd and sge_qmaster. To add the services to your NIS map, choose reserved, unused port numbers. The following examples show sge_qmaster and sge_execd entries. sge_qmaster 6444/tcp sge_execd 6445/tcp Installation Methods Several methods are available for installing the Grid Engine software: Interactive Interactive, with increased security Planning the Installation 1-7

18 Directory Organization Automated, using the inst_sge script and a configuration file Upgrade To decide which installation method you should use, consider the following factors. Do you already have the Grid Engine software installed and running? If so, you will probably want to upgrade. The upgrade process is described in Upgrading Grid Engine. If not, the master host installation is only done once. The master host is typically installed interactively, as described in Installing the Software With the GUI Installer or Installing the Software From the Command Line. Do you need to install just a few execution hosts? If so, then you will probably want to install them interactively, as described in Installing the Software With the GUI Installer or Installing the Software From the Command Line. Do you need to install a large number of execution hosts? If so, then you might want to perform automated installation, using the inst_sge script and a configuration file. See Using the inst_sge Utility and a Configuration Template. Do you require your grid to use encryption? If so, you have to perform an interactive installation with increased security. See Installing the Increased Security Features. Directory Organization When determining the directory organization, you must decide the following: The directory organization, for example, whether you will install a complete software tree on each workstation, cross-mounted directories, or a partial directory tree on some workstations. Where to locate each $SGE_ROOT root directory. Note: Because changing the installation directory or the spool directories requires a new installation of the system, use extra care to select a suitable installation directory. You can preserve all important information from a previous installation. By default, the installation procedure installs the Grid Engine software, man pages, spool areas, and the configuration files in a directory hierarchy under the installation directory as shown in the following figure. If you accept this default behavior, you should install or select a directory with the access permissions that are described in File Access Permissions. 1-8 Oracle Grid Engine Installation and Upgrade Guide

19 Directory Organization Figure 1 1 Sample Directory Hierarchy You can choose to put the spool areas in other locations during the primary installation. See Oracle Grid Engine Administration Guide for more detailed instructions about configuring queues. Spool Directories under the Root Directory During the installation of the master host, you must specify the location of a spooling directory. This directory is used to spool jobs from execution hosts that do not have a local spooling directory. Note: If you are using a Windows execution host, you must use the local spooling directory. On the master host, spool directories are maintained under qmaster-spool-dir. The location of qmaster-spool-dir is defined during the master host installation process. The default value of qmaster-spool-dir is $SGE_ROOT/$SGE_ CELL/spool/qmaster. On each execution host, a spool directory called execd-spool-dir is defined during the execution host installation processes. The default value of execd-spool-dir is $SGE_ROOT/$SGE_CELL/spool/exec-host. You will get better performance from execution hosts with local spooling directories than from execution hosts that have NFS mounted the master host's spooling directory. Planning the Installation 1-9

20 Directory Organization Note: If no cell name is specified during installation, the value of $SGE_CELL is default. You do not need to export these directories to other machines. However, exporting the entire $SGE_ROOT tree and making it write-accessible for the master host and all executable hosts makes administration easier. Note: If you use a Lustre fileshare as the spool directory, you should disable file striping for these directories. For information about how to disable file striping, refer to the Lustre operation manual located at: Choosing Between Classic Spooling and Database Spooling $SGE_ROOT Directory During the installation, you are given the option to choose between classic spooling and Berkeley DB spooling. If you choose Berkeley DB spooling, you are then given the option to spool to a local directory or to a separate host, known as a Berkeley DB spooling server. Using a Berkeley DB spooling server might provide better performance than classic spooling. Part of this performance increase is because the master host can make non-blocking writes to the database, but has to make blocking writes to the text file used by classic spooling. Also consider file format and data integrity. Writing to the Berkeley DB provides a greater level of data integrity than writing to a text file. However, a text file stores data in a format that you can read and edit. Normally, you do not need to read these files, but the spooling directory contains the messages from the system daemons, which can be useful for debugging. You must create a directory into which to load the contents of the distribution media. This directory is called the root directory, or $SGE_ROOT. When the Grid Engine system is running, this directory stores the current cluster configuration and all other data that must be spooled to disk. Note: For efficient spooling, place the spooling directories somewhere other than within $SGE_ROOT. Use a valid path name for the directory that is network-accessible on all hosts. For example, if the file system is mounted using automounter, set $SGE_ROOT to /usr/sge6, not to /tmp_mnt/usr/sge6. Note: Throughout this information space, the $SGE_ROOT environment variable is used to refer to the directory into which the Grid Engine software is installed. The $SGE_ROOT directory is the top level of the Grid Engine software directory tree. On startup, each Grid Engine software component in a cell needs read access to the $SGE_ ROOT/$SGE_CELL/common directory. When Grid Engine software is installed as a single cluster, the value of $SGE_CELL is default Oracle Grid Engine Installation and Upgrade Guide

21 Spooling Options For ease of installation and administration, this directory should be readable on all hosts on which you intend to run the Grid Engine software installation procedure. For example, you can select a directory that is available across a network file system, such as NFS. If you choose to select file systems that are local to the hosts, you must copy the installation directory to each host before you start the installation procedure for the particular machine. See File Access Permissions for a description of required permissions. Spooling Options During the installation, you are given the option to choose between classic spooling and Berkeley DB spooling. If you choose Berkeley DB spooling, you are then given the option to spool to a local directory or to a separate host, known as a Berkeley DB spooling server. Using a Berkeley DB spooling server might provide better performance than classic spooling. Part of this performance increase is because the master host can make non-blocking writes to the database, but has to make blocking writes to the text file used by classic spooling. Also consider file format and data integrity. Writing to the Berkeley DB provides a greater level of data integrity than writing to a text file. However, a text file stores data in a format that you can read and edit. Normally, you do not need to read these files, but the spooling directory contains the messages from the system daemons, which can be useful for debugging. Database Server and Spooling Host The master host can store its configuration and state to a Berkeley DB spooling database. The spooling database can be installed on the master server or on a separate host. When the Berkeley DB spools into a local directory on the master host, the performance is better. If you want to set up a shadow master host, you need to use a separate Berkeley DB spooling server (host). In this case, you have to choose a host with a configured RPC service. The master host connects through RPC to the Berkeley DB. Note: This configuration does not provide a High-Availability (HA) solution. For example, scripts of pending jobs are not spooled through BDB spool server and thus are not available for a shadow master. With the introduction of NFS4 software available with the Solaris 10 operating system, you can use Berkeley DB spooling on a network file system. You could not use Berkeley DB spooling on previous NFS versions. This circumstance allows a shadow host installation spooled on Berkeley DB without setting up an additional Berkeley DB Spooling Server. Caution: Although using a shadow master host is more reliable, using a separate Berkeley DB spooling host results in a potential security hole. RPC communication as used by the Berkeley DB can be easily compromised. Only use this alternative if your site is secure and if users can be trusted to access the Berkeley DB spooling host by means of TCP/IP communication. If you choose to use Berkeley DB spooling without a shadow master, you do not need to set up a separate spooling server. Likewise, if you choose not to use Berkeley DB Planning the Installation 1-11

22 Scheduler Profiles Scheduler Profiles spooling, you can set up a shadow master host without setting up a separate spooling server. Once you determine whether you need a separate spooling server, you will also need to determine the location for the spooling directory. The spooling directory must be local to the spooling server. A default value for the location of the spooling directory is recommended during installation, but this default value is not suitable when the file server is different from the master host. The requirements for the Berkeley DB spooling host are similar to the requirements for the master host: The host must be a stable platform. The host must not be excessively busy with other processing. At least 60 to 120 Mbytes of unused main memory must be available to run the Grid Engine system daemons. For very large clusters that include many hundreds or thousands of hosts and tens of thousands of jobs in the system at any time, one GByte or more of unused main memory might be required and two CPUs might be beneficial. (Optional) A separate spooling host must be installed before the master host. (Optional) The $SGE_ROOT directory should be installed locally, to cut down on network traffic. You can choose from three scheduler profiles during the installation process: normal, high, and max. You can use these predefined profiles as a starting point for Grid Engine tuning. Using these profiles, you can optimize the scheduler for one or more of the following: The amount of information that is tracked about a scheduling run The load adjustment during a scheduling run Interval scheduling (the default) or immediate scheduling You can choose from three scheduler profiles: normal - This profile uses load adaptation and interval scheduling, and reports all the information that the scheduler gathers during the dispatch cycle. This profile is the starting point for most grids. Use this profile if your highest priority is gathering and reporting information about a scheduling run. high - This profile is more appropriate for a large cluster, where throughput is more important than gathering and reporting all the information from the scheduler. This profile also uses interval scheduling. Use this profile if you want to get better performance at the cost of getting less information about your scheduling runs. max - This profile disables all information gathering and reporting, enables immediate scheduling, and disables load adaptation. Immediate scheduling is very useful for sites with high throughput and very short running jobs. The advantage of immediate scheduling decreases as runtime of the jobs increases. This profile can be used in clusters of any size where only throughput is important and everything else is a lower priority. For more information on how to configure scheduling, see Oracle Grid Engine Administration Guide Oracle Grid Engine Installation and Upgrade Guide

23 Getting the Software Getting the Software Electronic Download CD-ROM Distribution The software is distributed through electronic download and on CD-ROM. To electronically download a copy of the Grid Engine software, visit SUN.COM. The product distribution is in pkgadd format for the Solaris Operating System (Solaris OS). If you would like to download a copy of the open source Grid Engine software, visit the download center. For information on how to access CD-ROMs, ask your system administrator or refer to your local system documentation. For instructions, see Loading the Distribution Files on a Workstation. Planning the Installation 1-13

24 Getting the Software 1-14 Oracle Grid Engine Installation and Upgrade Guide

25 2 2Installing Grid Engine To effectively install Grid Engine, perform the following tasks in the order that they are listed: Table 2 1 Topic Installation Tasks Planning the Installation Loading the Distribution Files on a Workstation Installing the Software With the GUI Installer Installing the Software From the Command Line Installing the Increased Security Features Oracle Grid Engine User s Guide Verifying the Installation Description Strategically plan your installation to achieve results that fit your environment. Unpack and load the distribution files onto a workstation. Learn how to run the new GUI installer and install whole cluster. Learn how to run an installation script on the master host and on every execution host in the Grid Engine system and to register information about administration hosts and submit hosts. Set up your system more securely. Install the Accounting and Reporting Console, an optional feature that enables you to gather live reporting data from the Grid Engine system. Verify that the daemon is running on the master host and on the Execution Hosts and how to run simple commands and submit test jobs. In addition, you might need to perform one or more related tasks: Table 2 2 Topic Additional Installation Tasks Automating the Installation Process Installing SMF Services Installing a JMX-Enabled System Removing the Software Additional Software for the Microsoft Operating System Description Learn how to automate the Grid Engine installation process. Learn how to install the Service Management Facility (SMF) services. Learn how to install a JMX-enabled system. Learn how to remove the Grid Engine software. Learn how to install Grid Engine on Microsoft Windows operating system. Installing Grid Engine 2-1

26 Loading the Distribution Files on a Workstation Table 2 2 (Cont.) Additional Installation Tasks Topic User Management on Windows Hosts Other Installation Issues Description Learn how to manage user accounts on Windows hosts. Learn how to identify additional considerations for installing Grid Engine software. Loading the Distribution Files on a Workstation The Grid Engine 6.2 software is distributed on CD-ROM and through electronic download. The CD-ROM distribution contains a directory named Sun_Grid_Engine_ 6_2. The product distribution is in this directory, in both tar.gz format and the pkgadd format. The pkgadd format is provided for the Solaris Operating System (Solaris OS). For all supported operating systems, the software is distributed in tar.gz format. For more on how to obtain the distribution files, see Getting the Software. How to Load the Distribution Files on a Workstation Ensure that the file systems and directories that are to contain the Grid Engine software distribution and the spool and configuration files are set up properly by setting the access permissions as defined in File Access Permissions. 1. Provide access to the distribution media. If you downloaded the software, rather than getting it on CD-ROM, just unzip the files into a directory. This directory must be located on a file system that has at least 350 MBytes free space. 2. Log in to a system. Log in preferably on a system that has a direct connection to a file server. 3. Create the installation directory. Create an installation directory as described in $SGE_ROOT Directory. # mkdir /opt/sge6-2 In these instructions, the installation directory is abbreviated as sge-root. 4. Install the binaries for all binary architectures that are to be used by any of your master, execution, and submit hosts in your Grid Engine system cluster. You can use either the pkgadd Method or the tar Method. pkgadd Method The pkgadd format is provided for the Solaris Operating System. To facilitate remote installation, the pkgadd directories are also provided in zip files. You can install the following packages: Table 2 3 Package SUNWsgeec Installing Packages Using Pkgadd Method Description Architecture independent files SUNWsgeex Solaris (SPARC platform) 64-bit binaries for Solaris 8, Solaris 9, and Solaris 10 Operating Systems SUNWsgeei SUNWsgeeax Solaris (x86 platform) binaries for Solaris 8, Solaris 9, and Solaris 10 Operating Systems Solaris (x64 platform) binaries for Solaris 10 Operating System 2-2 Oracle Grid Engine Installation and Upgrade Guide

27 Loading the Distribution Files on a Workstation Table 2 3 (Cont.) Installing Packages Using Pkgadd Method Package SUNWsgeea Description Accounting and Reporting Console (ARCo) packages for the Solaris and Linux Operating systems. As you type the following commands, you must be prepared to respond to script questions about your base directory, sge-root, and the administrative user. The script requests the choices that you made during the planning steps of this installation. See Planning the Installation for further details. At the command prompt, type the following commands, responding to the script questions. # cd cdrom_mount_point/sun_grid_engine_6_2 # pkgadd -d./common/packages SUNWsgeec Depending on the Solaris binary that you need, type one of the following commands: # pkgadd -d./solaris_sparc/packages SUNWsgee # pkgadd -d./solaris_sparc/packages SUNWsgeex # pkgadd -d./solaris_x86/packages SUNWsgeei # pkgadd -d./solaris_x64/packages SUNWsgeeax tar Method For all supported operating systems, the software is distributed in tar.gz format. Regardless of platform, install the architecture independent file Common/tar/sge-6_ 2-common.tar.gz. The tar files that contain platform-specific binaries use the naming convention of sge-6_2-bin-architecture.tar.gz. The following table lists the platform-specific binaries. Install the file for each platform that you need to support. Note that each platform has its own directory under Sun_ Grid_Engine_6_2. Table 2 4 Platform-Specific File Installing Binaries Using Tar Method Solaris_sparc/tar/sge-6_ 2-bin-solaris-sparcv9.ta r.gz Solaris_x86/tar/sge-6_ 2-bin-solaris-i586.tar.g z Solaris_x64/tar/sge-6_ 2-bin-solaris-x64.tar.gz Windows/tar/sge-6_ 2-bin-windows-x86.tar.gz Linux24_i586/tar/sge-6_ 2-bin-linux24-i586.tar.g z Linux24_amd64/tar/sge-6_ 2-bin-linux24-ia64.tar.g z Platform Solaris (SPARC platform) 64-bit binaries for Solaris 8, Solaris 9, and Solaris 10 Operating Systems Solaris (x86 platform) binaries for Solaris 8, Solaris 9, and Solaris 10 Operating Systems Solaris (x64 platform) 64-bit binaries for Solaris 10 Microsoft Windows (x86 platform) 32-bit binaries for Windows 2000, XP and Windows Server 2003 Linux (x86 platform) binaries for the 2.4 and 2.6 kernel Linux (Itanium platform) binaries for the 2.4 and 2.6 kernel Installing Grid Engine 2-3

28 Installing the Software With the GUI Installer Table 2 4 (Cont.) Installing Binaries Using Tar Method Platform-Specific File Linux24_amd64/tar/sge-6_ 2-bin-linux24-x64.tar.gz MacOSX/tar/sge-6_ 2-bin-darwin-ppc.tar.gz MacOSX/tar/sge-6_ 2-bin-darwin-x64.tar.gz HPUX11/tar/sge-6_ 2-bin-hp11.tar.gz HPUX11/tar/sge-6_ 2-bin-hp11-64.tar.gz Aix43/tar/n1ge-6_ 1-bin-aix51.tar.gz Platform Linux binaries for the 2.4 and 2.6 kernel Apple Mac OS/X (PowerPC platform) Apple Mac OS/X (Intel-based platform) Hewlett-Packard HP-UX 11 or higher 64-bit binaries for Hewlett-Packard HP-UX 11 or higher IBM AIX 5.1 and 5.3 Type the following commands at the command prompt. In the example, <basedir> is the abbreviation for the full directory, cdrom-mount-point/sun_grid_engine_6_2. % su # cd <sge-root> # gzip -dc <basedir>/common/tar/sge-6_2-common.tar.gz tar xvpf - # gzip -dc <basedir>/solaris_sparc/tar/sge-6_2-bin-solsparc32.tar.gz tar xvpf - # gzip -dc <basedir>/solaris_sparc/tar/sge-6_2-bin-solsparc64.tar.gz tar xvpf - # SGE_ROOT=<sge-root>; export SGE_ROOT # util/setfileperm.sh $SGE_ROOT Installing the Software With the GUI Installer A new GUI installer to simplify the installation process is available since Grid Engine 6.2u2. The GUI installer enables you to easily install a whole cluster interactively. To install a cluster, you need to set up the environment in a similar way to an automatic installation. Requirements The GUI installer requires at least Version 5 of the Java platform. Screen resolution of 1024x768 or larger. (Optional) Password-less ssh or rsh access as root user to all remote hosts that you want to install. If this requirement is not met you can only install Grid Engine components on a local host. For more information, see How to Configure Password-less Access for the root User. You can still use the GUI installer by starting it locally from each remote host. Start the installer as root user. Ensure that you start the installation from the qmaster host when password-less root access is available. For information on installation modes supported by the GUI installer, see these topics: 2-4 Oracle Grid Engine Installation and Upgrade Guide

29 Express Installation Topic Express Installation Custom Installation Description Enables first-time users to install the software easily. Provides a significantly reduced set of parameters that need to be configured. Requires password-less ssh access as root user to all remote hosts that you want to install. Enables you to configure almost all existing options that are available during the command-line installation. Offers more advanced features for the cluster host selection. Requires password-less ssh or rsh access as root user to all remote hosts that you want to install. For additional reference information, see these topics: Topic How to Configure Password-less Access for the root User Understanding Host and Installation States Tweaking start_gui_ installer Troubleshooting the GUI Installer Express Installation The express installation mode is targeted at first-time users and provides a significantly reduced set of parameters to configure. This mode also provides reasonable default values for most of the parameters. You must have a password-less ssh or scp access if you are planning to install Grid Engine on remote hosts. The following steps describe a complete cluster installation and assume that the password-less access is configured. (Click any of the screen captures in the following steps to view more details.) Using the Express Installation Mode Description Procedure for configuring a password-less ssh or rsh access for the root user to install a whole Grid Engine (SGE) cluster by using the GUI installer. Describes the different installation states that you might encounter while using the GUI installer. Describes the command-line options of the start_gui_installer command and how to use them to fine tune the performance of the installer. Contains known issues and their workarounds. The express installation steps are as follows. 1. Start the GUI installer. On the welcome screen, click Next. Note: Ensure that you start the GUI installer on the qmaster host. As root, run the start_gui_installer command in your sge-root directory. For example: master:/sge#./start_gui_installer Starting Installer Choose components to install. Click Next. See the following table for a brief explanation of options displayed on this screen. Installing Grid Engine 2-5