README VERSION : 1.1 README CREATION DATE : 2014-07-10 PATCH-ID : VRTSvcs-6.1.1.000 PATCH NAME : VRTSvcs 6.1.1.000 BASE PACKAGE NAME : VRTSvcs BASE PACKAGE VERSION : 6.1.0.000 SUPERSEDED PATCHES : NONE REQUIRED PATCHES : NONE INCOMPATIBLE PATCHES : NONE SUPPORTED PADV : sol11_sparc (P-PLATFORM , A-ARCHITECTURE , D-DISTRIBUTION , V-VERSION) PATCH CATEGORY : CORE , CORRUPTION PATCH CRITICALITY : CRITICAL HAS KERNEL COMPONENT : NO ID : NONE REBOOT REQUIRED : NO REQUIRE APPLICATION DOWNTIME : YES PATCH INSTALLATION INSTRUCTIONS: -------------------------------- Please refer to Install Guide for install instructions and Release Notes for other information. PATCH UNINSTALLATION INSTRUCTIONS: ---------------------------------- Please refer to Install Guide for uninstall instructions and Release Notes for other information. SPECIAL INSTRUCTIONS: --------------------- NONE SUMMARY OF FIXED ISSUES: ----------------------------------------- PATCH ID:VRTSvcs-6.1.1.000 3389650 (3389647) The hazonesetup and hazoneverify utilities do not support multiple zone resources in the same service group. 3453222 (3436617) VCS fails to execute triggers. 3472613 (3471819) The service group fails to go online if the CurrentCount attribute value is incorrect. 3476947 (3385820) Sometimes the high availability daemon (HAD) crashes if it runs for a long duration. 3485944 (3485941) A service group gets stuck in the STARTING state. 3499034 (3498072) The hazonesetup(1M) utility reports a Perl warning message when the locale is set to non-English. 3508686 (3381042) Unexpected deletion of temporary files during VCS start causes the VCS agents to report an incorrect state. SUMMARY OF KNOWN ISSUES: ----------------------------------------- NONE KNOWN ISSUES : -------------- NONE FIXED INCIDENTS: ---------------- PATCH ID:VRTSvcs-6.1.1.000 * INCIDENT NO:3389650 TRACKING ID:3389647 SYMPTOM: The hazonesetup and hazoneverify utilities do not support multiple zone resources in the same service group. DESCRIPTION: ResContainerInfo is a resource level attribute, which is introduced to support multiple zone resources in the same service group. However, the hazonesetup and hazoneverify utilities fail to support the ResContainerInfo attribute. RESOLUTION: The code of the hazonesetup and hazoneverify utilities is modified to support multiple zone resources in the same service group. * INCIDENT NO:3453222 TRACKING ID:3436617 SYMPTOM: VCS fails to execute triggers. DESCRIPTION: If some arguments are left unassigned while invoking triggers, the hatrigger command fails due to compilation errors. RESOLUTION: The code is modified to combine all arguments in a single string before invoking the command. * INCIDENT NO:3472613 TRACKING ID:3471819 SYMPTOM: The service group fails to go online if the CurrentCount attribute value is incorrect. DESCRIPTION: Whenever a resource goes online, the CurrentCount attribute value is incremented. When the CurrentCount value is greater than 1, the concurrency violation trigger is invoked. VCS internally calls flush on the service group which clears the IState for the resources and also decrements the CurrentCount value. However, due to an incorrect update of the CurrentCount attribute, its value gets incorrectly updated. As a result, subsequent online of the service group fails with the following error message: VCS WARNING V-16-1-10165 Failover group is not completely offline in cluster RESOLUTION: The code to set the CurrentCount attribute value is modified so that the attribute value is correctly set after concurrency violation. * INCIDENT NO:3476947 TRACKING ID:3385820 SYMPTOM: Sometimes the high availability daemon (HAD) crashes if it runs for a long duration. DESCRIPTION: VCS communicates with clients over Inter Process Messaging (IPM) channel. VCS uses an internal variable to keep count of the total number of connections opened by clients. When this value exceeds its maximum positive limit, HAD process crashes due to invalid memory access. RESOLUTION: The code is modified to correctly handle the limit of the internal variable. * INCIDENT NO:3485944 TRACKING ID:3485941 SYMPTOM: When you try to bring a service group online, the service group may get stuck in the STARTING state. Any attempt to bring resources within the service group online may also fail. DESCRIPTION: When the High Availability Daemon (HAD) tries to initiate the online operation for the service group, if a non-critical resource in the service group has faulted, the above RESOLUTION: The code is modified to ensure that if any non-critical resource is in the FAULTED state, HAD does not initiate an online operation for the service group. * INCIDENT NO:3499034 TRACKING ID:3498072 SYMPTOM: The hazonesetup(1M) utility reports a Perl warning message when the locale is set to non-English. DESCRIPTION: The hazonesetup(1M) utility runs the 'ha' command to extract the node ID. If the locale is set to non-English, the command fails to extract the node ID and logs a Perl warning message. RESOLUTION: The code is modified to set the environment variable LANG to C, extract the node ID, and reset the LANG variable. * INCIDENT NO:3508686 TRACKING ID:3381042 SYMPTOM: The checkboot utility core dump or time difference between a system and Network Time Protocol (NTP) time leads to unexpected deletion of the temporary files. The deletion causes the VCS agents to report an incorrect state. DESCRIPTION: The utmp(5M) file (file names are based on the operating system) contains a record of the reboots performed on a system. The hastart(1M) command uses the checkboot utility, which uses the functions provided by the operating system. The operating systems in response use the utmp file to identify the systems that are rebooted so that the temporary files for various agents are deleted before the agent startup. The reboot check is performed when you run the commands like hastart-version rather than before the High Availability Daemon (HAD) is started. If the utmp file is deleted, the checkboot utility fails due to a NULL pointer dereference, resulting in the deletion of the temporary files. The difference between the system time and the time provided by NTP server may also lead to an unexpected deletion of these temporary files. Thus, some of the agents which depend on these temporary files fail. RESOLUTION: The code is modified in the hastart(1M) command and checkboot utility so that the checkboot utility is appropriately invoked and error conditions are correctly handled. INCIDENTS FROM OLD PATCHES: --------------------------- NONE