README VERSION : 1.1 README CREATION DATE : 2012-09-18 PATCH-ID : 145452-04 PATCH NAME : VRTSdbac 5.1 SP1RP3 BASE PACKAGE NAME : Veritas Oracle Real Application Cluster Support Package by Symantec BASE PACKAGE VERSION : VRTSdbac 5.1SP1 SUPERSEDED PATCHES : NONE REQUIRED PATCHES : NONE INCOMPATIBLE PATCHES : NONE SUPPORTED PADV : sol10_x86 (P-PLATFORM , A-ARCHITECTURE , D-DISTRIBUTION , V-VERSION) PATCH CATEGORY : OTHER PATCH CRITICALITY : OPTIONAL HAS KERNEL COMPONENT : NO ID : NONE REBOOT REQUIRED : NO PATCH INSTALLATION INSTRUCTIONS: -------------------------------- Please refer the release notes for installation instructions PATCH UNINSTALLATION INSTRUCTIONS: ---------------------------------- Please refer the release notes for un-installation instructions. SPECIAL INSTRUCTIONS: --------------------- NONE SUMMARY OF FIXED ISSUES: ----------------------------------------- 2729806 (2853859) MultiPrivNIC on zonesg is plumbing private network interfaces on global zones 2740150 (2933837) SFRAC CPI does not set OfflineWaitLimit attribute for CSSD agent resource. 2746948 (2746940) Sometimes during bootup, some of the drivers fail to add in the system because of add_drv/rem_drv race between our modules which are independent of each other. SUMMARY OF KNOWN ISSUES: ----------------------------------------- KNOWN ISSUES : -------------- FIXED INCIDENTS: ---------------- PATCH ID:145452-04 * INCIDENT NO:2729806 TRACKING ID:2853859 SYMPTOM: MultiPrivNIC on zonesg is plumbing private network interfaces on global zones. DESCRIPTION: Logic to plumb private interfaces inside the zone was incorrect. RESOLUTION: Fixed the logic, now if zone is running we plumb private interfaces inside the zone. * INCIDENT NO:2740150 TRACKING ID:2933837 SYMPTOM: When CPI configures the CSSD resource, CSSD resource goes in ONLINE|UNABLE TO OFFLINE state after first monitor interval cycle. DESCRIPTION: When CPI configures the CSSD resource, CPI does not set OfflineWaitLimit=3 during modifying the CSSD resource attributes RESOLUTION: CPI will set OfflineWaitLimit=3 during configuring the CSSD resource by >>hares -override cssd_resname OfflineWaitLimit >>hares -modify cssd_resname OfflineWaitLimit 3. * INCIDENT NO:2746948 TRACKING ID:2746940 SYMPTOM: Sometimes during bootup, some of the drivers fail to add in the system because of add_drv/rem_drv race between our modules which are independent of each other. DESCRIPTION: When SMF framework start adding/removing drivers, sometimes race happens between multiple drivers and causes operation failure. RESOLUTION: Symantec is handling driver add/remove failure by adding retry logic in driver specific SMF script. INCIDENTS FROM OLD PATCHES: --------------------------- Patch Id::145452-03 * Incident no::2374970 Tracking ID ::2210343 Symptom::CRSResource agent does not support Oracle 11gR2. Description::CRSResource agent support for Oracle 11gR2. Resolution::CRSResource agent support for Oracle 11gR2. * Incident no::2374977 Tracking ID ::2231603 Symptom::If a node running SFRAC crashes and comes back online, Oracle instance on any other existing in the live cluster might crash throwing the IPC errors in the error logs. Description::As per the IPC implementation, if a node running SFRAC crashes and comes back online, Oracle instance on any other existing in the live cluster might crash throwing the IPC errors in the error logs. Resolution::Perform the cancellation of inprogress rqhs for VSND rqhs too in vcsipc_destwr() to avoid memory corruption issues during flow-control and peer node death. * Incident no::2380469 Tracking ID ::2551431 Symptom::SFRAC does not have support for Solaris local zones. Description::SFRAC cannot run in Solaris local zone environments where SFRAC is installed and configured on global zones while Oracle RAC is installed inside local zones. Resolution::Now SFRAC can run in Solaris local zone environments where SFRAC is installed and configured on global zones while Oracle RAC is installed inside local zones. * Incident no::2390892 Tracking ID ::2390967 Symptom::Starting VCSMM driver on one or more nodes in the cluster causes a memory leak in the vcsmm_set_cluster_proto function during memory allocation. Description::If VCSMM driver is started on one or more nodes in the cluster, it causes memory leaks. The amount of memory leak depends on the number of nodes in the cluster. Once the nodes join the cluster, starting VCSMM on any node in the cluster causes a further increase in the memory leak. Resolution::The vcsmm_set_cluster_proto function is now updated to allocate memory only if there is no current allocation. * Incident no::2429449 Tracking ID ::2511099 Symptom::Monitor script for cssd agent (ie. cssd-monitor) uses hard-coded name for cssd resource and due to this, if resource name used in main.cf is not "cssd", weird behavior is shown. Description::If cssd resource name in main.cf is not given as "cssd", cssd-monitor script fails to provide expected behavior. Resolution::Monitor script for cssd resource agent (ie. cssd-monitor) now handles the case where name of the resource is different. It retrieves the name and uses that instead of the hard-coded string "cssd".