* * * READ ME * * * * * * CPI 6.0.1 * * * * * * Public Hot Fix 8 * * * Patch Date: 2013-11-07 This document provides the following information: * PATCH NAME * OPERATING SYSTEMS SUPPORTED BY THE PATCH * PACKAGES AFFECTED BY THE PATCH * BASE PRODUCT VERSIONS FOR THE PATCH * SUMMARY OF INCIDENTS FIXED BY THE PATCH * DETAILS OF INCIDENTS FIXED BY THE PATCH * INSTALLATION PRE-REQUISITES * INSTALLING THE PATCH * REMOVING THE PATCH PATCH NAME ---------- CPI 6.0.1 Public Hot Fix 8 OPERATING SYSTEMS SUPPORTED BY THE PATCH ---------------------------------------- BASE PRODUCT VERSIONS FOR THE PATCH ----------------------------------- * Veritas Cluster Server 6.0.1 * Veritas Storage Foundation for Oracle RAC 6.0.1 * Veritas Storage Foundation Cluster File System 6.0.1 * Veritas Storage Foundation 6.0.1 * Veritas Storage Foundation High Availability 6.0.1 * Veritas Storage Foundation Basic 6.0.1 * Veritas Dynamic Multi-Pathing 6.0.1 * Symantec VirtualStore 6.0.1 * Veritas Storage Foundation for Sybase ASE CE 6.0.1 SUMMARY OF INCIDENTS FIXED BY THE PATCH --------------------------------------- Patch ID: 6.0.1.800 * 2984386 (3248893) Upon installation of SF version 6.0.1 getting the following: WARNING V-9-40-3995 Solaris 10 Update 6 is installed on target host. It is strongly recommended to install Solaris 10 Update 8 or later * 3125432 (3125494) Reporting $padv->QEMU not defined. * 3227558 (3243089) During a live upgrade the installation process takes more time than expected. * 3228679 (3230218) The Common Product Installer (CPI) based installation fails on Solaris version 11 because the nslookup command is missing. * 3294945 (3295841) CPI patches 6.0.1_P4.pl and 6.0.3_P6.pl fails when ssh banner is enabled * 3298901 (3304955) CPI shouldn't block upgrade if there's no configuration, it should at least upgrade all the packages. This is fixed in 6.1 and needs to be ported to 6.0.1 * 3307508 (3309409) Upon installation SFHA with any hotfix in Solaris 10 Update 11, installer getting the warning message about unsupported OS version. * 3360781 (3355748) Upgrade fails with multiple local zones in a service group DETAILS OF INCIDENTS FIXED BY THE PATCH --------------------------------------- This patch fixes the following Symantec incidents: Patch ID: 6.0.1.800 * 2984386 (Tracking ID: 3248893) SYMPTOM: Installer fails to detect the OS update level if the Solaris 10 kernel patch is upgraded to Update 10, and issues the below warning message: WARNING V-9-40-3995 Solaris 10 Update 6 is installed on . It is strongly recommended to install Solaris 10 Update 8 or later DESCRIPTION: Solaris 10 u10 haven't got published when 6.0.1 released. CPI need to support the lastest kernel version during precheck procedure. RESOLUTION: CPI add support to the release Solaris 10 u10. * 3125432 (Tracking ID: 3125494) SYMPTOM: CPI fails with QEMU undefined when user have bios vendor QEMU. DESCRIPTION: If customer have bios vendor QEMU, CPI will report the following error message: CPI ERROR V-9-0-0 1 thread 1 threw error: Can't locate object method "QEMU" via package "Padv::RHEL5x8664" RESOLUTION: CPI replace $padv->QEMU by $padv->VIRT_QEMU in padv.pm. * 3227558 (Tracking ID: 3243089) SYMPTOM: During a live upgrade on SunOS, the rsh or the ssh command use approximately 3 minutes per node to verify if a node is alive. The verification process causes a delay in the installation process. DESCRIPTION: During an installation process, when the rsh or the ssh command is invoked, the command verifies the state of each node to understand if it is alive. In the process of verification, the commands consume more time than expected, which causes a delay in the installation process. The installer also reports error messages during the rsh communication precheck. Some of these error messages are invalid that leads to repeated verification of the ssh or the rsh connection. RESOLUTION: Remove the invalid error messages while CPI verifies the rsh or the ssh commands. CPI executes the rsh or the ssh communication only once regardless of the failure that is observed in the rsh or the ssh verification process. * 3228679 (Tracking ID: 3230218) SYMPTOM: With Solaris version 11, the Common Product Installer (CPI) based installation process fails during a precheck because the nslookup command is missing. After the failure, the installer reports the following error message: CPI ERROR V-9-30-1446 The following required OS commands are missing on system: /usr/sbin/nslookup: No such file or directory. DESCRIPTION: The Internet Systems Consortium (ISC) has deprecated the use of nslookup command instead has certified to use either the host or dig. The CPI packages contain the nslookup command, which the standard build does not support. During an installation, the Storage Foundation (SF) CPI installer installs the bind packages (pkg:/network/dns/bind) and later removes these bind packages to comply with the ISC's new security guidelines. However, during a precheck when the nslookup command is not found, the installer stops the installation with an error. RESOLUTION: To prevent the installation failure, remove the nslookup command from the preinstallation command checklist. * 3294945 (Tracking ID: 3295841) SYMPTOM: During an installation process with 6.0.1_P4.pl or 6.0.3_P6.pl, when ssh banner is enabled, an error will be given: Checking system communication CPI ERROR V-9-0-0 1 thread 1 threw error: Can't locate object method "arch_sys" via package "Padv::" (perhaps you forgot to load "Padv::"?) at /opt/VRTSperl/lib/site_perl/UXRT60/EDR/EDR/systems.pm line 102. DESCRIPTION: During an installation process, when the rsh or the ssh command is invoked, the command verifies the state of each node to understand if it is available to conection. If ssh/rsh banner is enabled, the existing diagnostic method for connection will failed. RESOLUTION: Installer will not set $sys->{banner} when testing ssh/rsh connection. * 3298901 (Tracking ID: 3304955) SYMPTOM: Installer block upgrade to 6.0.1 if no VCS configuration exist. DESCRIPTION: T-Mobile like many customers has a "Gold Image" install methodology; all Servers get the same OS and Storage Foundation packages - in this case T-Mobile installs SFRAC. The issue here is on the many servers that are used in standalone mode with only Storage Foundation and DMP configured and running. The installer detects the SFRAC packages and that VCS is not configured and then stops running. Our customer must then manually uninstall various packages, which is not acceptable. RESOLUTION: Installer will not block upgrade, just give warnings and user can continue upgrade without VCS configuration. * 3307508 (Tracking ID: 3309409) SYMPTOM: Installation of SFHA 6.0.1 with CPI hotfix in Solaris 10 Update 11, installer gives the following warning messages: "The following warnings were discovered on the systems: CPI WARNING V-9-40-3995 Solaris 10 Update 6 is installed. It is strongly recommended to install Solaris 10 Update 8 or later." DESCRIPTION: 148888-03 is Solaris 10 Update 11 kernel patch ID which was released January 2013.While the first CPI HF for 6.0.1 was created last year, which to support Solaris10 U10, so Solaris 11 Update 11 support was not included. RESOLUTION: Add the Solaris 11 Update 11 support to the latest hotfix. * 3360781 (Tracking ID: 3355748) SYMPTOM: Upgrading VCS from 5.0MP3RP5 to 6.0.1 with SG containing multiple zones throws CPI errors CPI ERROR V-9-40-3737 In the file '/etc/VRTSvcs/conf/config/main.cf' on sprsv480b7-09, the ContainerName and ContainerType value of all resources in group 'SAP-HAWKER-ITE1-SG' are not consistent : SAP-HAWKER-ITE1-SG app-zone2-res ContainerName tstzone2 SAP-HAWKER-ITE1-SG app-zone2-res ContainerType Zone SAP-HAWKER-ITE1-SG app-zone1-res ContainerName tstzone1 SAP-HAWKER-ITE1-SG app-zone1-res ContainerType Zone CPI ERROR V-9-40-3738 In the file '/etc/VRTSvcs/conf/config/main.cf' on sprsv480b7-09, the group 'SAP-HAWKER-ITE1-SG' contains more than 1 Zone : SAP-HAWKER-ITE1-SG Zone Name tstzone1 SAP-HAWKER-ITE1-SG Zone Name tstzone2 DESCRIPTION: Since 6.0.1 VCS provide specific customer a hotfix to support multiple zones(via a new attribute ResContainerInfo in resource level) in a single service group as it's supported prior to 5.1. However CPI wasn't required to support this design change. So when the setup in such case upgrade from versions prior to 5.1 to 6.x, CPI throw error and bail out. RESOLUTION: Add the ResContainerInfo support for upgrading from versions prior to 5.1 to 6.x to support multiple zones in a single service group. INSTALLING THE PATCH -------------------- 1. Copy the hot-fix CPI_6.0.1_P8.pl to /tmp 2. Run installation scripts with the hot-fix Example:./installer -require /tmp/CPI_6.0.1_P8.pl REMOVING THE PATCH ------------------ 1. Copy the hot-fix CPI_6.0.1_P8.pl to /tmp 2. Run uninstall scripts with the hot-fix Example: ./installer -uninstall -require /tmp/CPI_6.0.1_P8.pl SPECIAL INSTRUCTIONS -------------------- NONE OTHERS ------ NONE