README VERSION : 1.1 README CREATION DATE : 2014-04-09 PATCH-ID : 6.0.500.000 PATCH NAME : VRTScavf 6.0.500.000 BASE PACKAGE NAME : VRTScavf BASE PACKAGE VERSION : 6.0.100.000 SUPERSEDED PATCHES : 6.0.300.000 REQUIRED PATCHES : NONE INCOMPATIBLE PATCHES : NONE SUPPORTED PADV : sol11_x64 (P-PLATFORM , A-ARCHITECTURE , D-DISTRIBUTION , V-VERSION) PATCH CATEGORY : OTHER PATCH CRITICALITY : CRITICAL HAS KERNEL COMPONENT : YES ID : NONE REBOOT REQUIRED : NO REQUIRE APPLICATION DOWNTIME : YES PATCH INSTALLATION INSTRUCTIONS: -------------------------------- Please refer to Install guide for install instructions PATCH UNINSTALLATION INSTRUCTIONS: ---------------------------------- Please refer to Install guide for uninstall instructions SPECIAL INSTRUCTIONS: --------------------- NONE SUMMARY OF FIXED ISSUES: ----------------------------------------- PATCH ID:6.0.500.000 3399337 (3410837) The error message has an issue when the user uses the cfsumount(1M) to unmount a mount point which has a samba parent resource. PATCH ID:6.0.300.000 2933754 (2715175) 4 node cluster shutdown time takes 30 minutes 2949033 (2942776) CFS mount issues/VxVM - Mount fails when volumes in vset is not ready SUMMARY OF KNOWN ISSUES: ----------------------------------------- 3306574(3313498) svsdatastore (1m) command may set the return value to zero even in cases of error. 3481929(3483394) It fails to add ClearClone Attribute automatically to CVMTypes.cf after upgrade KNOWN ISSUES : -------------- * INCIDENT NO::3306574 TRACKING ID ::3313498 SYMPTOM:: svsdatastore (1m) command may set the return value to zero even in cases of error. e.g. #svsdatastore add Error: V-35-585: Disk invaliddisk does not exists # echo $? 0 WORKAROUND:: None. * INCIDENT NO::3481929 TRACKING ID ::3483394 SYMPTOM:: WORKAROUND:: Manually add ClearClone attribute to the /etc/VRTSvcs/conf/config/CVMTypes.cf file as below: Replace old line with new line Old line: "static str ArgList[] = { CVMDiskGroup, CVMVolume, CVMActivation, CVMVolumeIoTest, CVMDGAction, CVMDeportOnOffline, CVMDeactivateOnOffline, State }" New line: "static str ArgList[] = { CVMDiskGroup, CVMVolume, CVMActivation, CVMVolumeIoTest, CVMDGAction, CVMDeportOnOffline, CVMDeactivateOnOffline, ClearClone, State }" Add the line of "int ClearClone" before that of "temp int voldg_stat". In case of rollback of 605, remove the added part from the file. FIXED INCIDENTS: ---------------- PATCH ID:6.0.500.000 * INCIDENT NO:3399337 TRACKING ID:3410837 SYMPTOM: The error message has an issue when the user uses the cfsumount(1M) to unmount a mount point which has a samba parent resource. DESCRIPTION: The cfsunmount(1M) script uses "hares --value" to get the mount point. If the mount point which has samba parent resource is still mounted, then Cfsunmount(1M) gives an error message such as "Error: V-35-201: Cannot unmount /cifsdg/datavol2 on l111053, dependent [ ] currently mounted on l111053". RESOLUTION: The error message is appropriately changed for a better judgment. PATCH ID:6.0.300.000 * INCIDENT NO:2933754 TRACKING ID:2715175 SYMPTOM: cfsumount command runs slowly on large file system, there are some file system reconfigure threads in the kernel. schedule vx_svar_sleep_unlock vx_event_wait vx_async_waitmsg vx_msg_broadcast vx_dele_rebuild vx_dele_assumption vx_recover vx_recover_evanesce vx_kthread_init kernel_thread DESCRIPTION: When unmounting a file system, the cfsumount will try to offline the file system on all nodes in parallel. If the primary file system node gets unmounted first, the process of reconfigure will be initiated to migrate the primary to other nodes, however, the reconfigure is unnecessary for cfsumount command. For large file systems, reconfigure process involves much work and may take a long time. RESOLUTION: Unmount the secondary nodes first to avoid unnecessary file system reconfigure. * INCIDENT NO:2949033 TRACKING ID:2942776 SYMPTOM: CFSmount is failing with the error ENXIO or EIO on volume vset device. The error messages look like UX:vxfs mount.vxfs: ERROR: V-3-20005: read of super-block on /dev/vx/dsk/sfsdg/vol1 failed: No such device or address or UX:vxfs mount.vxfs: ERROR: V-3-20005: read of super-block on /dev/vx/dsk/sfsdg/vol2 failed: Input/output error DESCRIPTION: If the volume set is in ENABLED state but some underlying volumes are not ready or in DISABLED state, the mount will fail with ENXIO or EIO. RESOLUTION: Code changes in cfsmntadm has been made to monitor the volume states in CVMVolDG agent. INCIDENTS FROM OLD PATCHES: --------------------------- NONE