Grid patching was failing with Failed to bring down CRS servic(Clean of ‘ora.gipcd’ on ‘host’ failed)
We were trying to patch out GI home with Oct 2020 PSU, but the patch was not proceeding and every time it failed , when it was trying to bring the CRS down
Executing patch validation checks on home /oracle/app/12.2.0/grid Patch validation checks successfully completed on home /oracle/app/12.2.0/grid Checking shared status of home..... Bringing down CRS service on home /oracle/app/12.2.0/grid Prepatch operation log file location: /oracle/app/grid/crsdata/host/crsconfig/crspatch_host_2021-02-10_09-02-31PM.log Failed to bring down CRS service on home /oracle/app/12.2.0/grid Execution of [GIShutDownAction] patch action failed, check log for more details. Failures: Patch Target : host->/oracle/app/12.2.0/grid Type[crs] CRS-4000: Command Start failed, or completed with errors. 2021/02/10 21:07:02 CLSRSC-117: Failed to start Oracle Clusterware stack
Then , we tried to manually bring the cluster down , to see where the issue is actually happening in the process of clean shutdown of CRS.
root@host (B)<-->{/oracle/app/12.2.0/grid/bin}:$ ./crsctl stop crs -f
CRS-2791: Starting shutdown of Oracle High Availability Services-managed resources on 'host'
CRS-2679: Attempting to clean 'ora.gipcd' on 'host'
CRS-2680: Clean of 'ora.gipcd' on 'host' failed
CRS-2799: Failed to shut down resource 'ora.gipcd' on 'host'
CRS-2795: Shutdown of Oracle High Availability Services-managed resources on 'host' has failed
CRS-4687: Shutdown command has completed with errors.
CRS-4000: Command Stop failed, or completed with errors.
root@host (B)<-->{/oracle/app/12.2.0/grid/bin}:$ ./crsctl stat res -t
CRS-4535: Cannot communicate with Cluster Ready Services
CRS-4000: Command Status failed, or completed with errors.
when we tried to forcefully shutdown the cluster , the we came to know that gipcd component of cluster was creating issue in clean shutdown of cluster .
We checked the gipcd status by
root@host (B)<-->{/oracle/app/12.2.0/grid/bin}:$ ./crsctl stat res -t -init
Name Target State Server State details
Cluster Resources
ora.asm
1 ONLINE OFFLINE STABLE
ora.cluster_interconnect.haip
1 ONLINE OFFLINE STABLE
ora.crf
1 ONLINE OFFLINE STABLE
ora.crsd
1 ONLINE OFFLINE STABLE
ora.cssd
1 ONLINE OFFLINE STABLE
ora.cssdmonitor
1 OFFLINE OFFLINE STABLE
ora.ctssd
1 ONLINE OFFLINE STABLE
ora.diskmon
1 OFFLINE OFFLINE STABLE
ora.drivers.acfs
1 OFFLINE OFFLINE STABLE
ora.evmd
1 ONLINE INTERMEDIATE host STABLE
ora.gipcd
1 ONLINE UNKNOWN host STABLE
ora.gpnpd
1 ONLINE ONLINE host STABLE
ora.mdnsd
1 ONLINE ONLINE host STABLE
ora.storage
1 ONLINE OFFLINE STABLE
The gipcd was in unknow state , was the reason , it was not allowing cluster to shut down properly
Checked all related logs to check , why it was in unknow state
/oracle/app/grid/diag/crs/host/crs/trace/ohasd.trc /oracle/app/grid/diag/crs/host/crs/trace/alert /oracle/app/grid/diag/crs/host/crs/trace/gipcd.trc
The gipcd trace was showing
2021-02-11 11:05:17.651 [OHASD(10617174)]CRS-5828: Could not start agent '/oracle/app/12.2.0/grid/bin/orarootagent'. Details at (:CRSAGF00123:) {0:0:2} in /oracle/app/grid/diag/crs/host/crs/trace/oh
asd.trc.
2021-02-11 11:05:17.651 [OHASD(10617174)]CRS-5828: Could not start agent '/oracle/app/12.2.0/grid/bin/orarootagent'. Details at (:CRSAGF00126:) {0:0:2} in /oracle/app/grid/diag/crs/host/crs/trace/oh
asd.trc.
2021-02-11 11:05:17.657 [OHASD(10617174)]CRS-2758: Resource 'ora.gipcd' is in an unknown state.
2021-02-11 11:05:17.688 [GPNPD(20644274)]CRS-2328: GPNPD started on node host.
Tried the below note CRS is not starting after applying the latest RU in 12.2 (Doc ID 2373945.1) , didn’t worked
crsctl disbale crs rebooted the node ( after reboot node will not AutoStart, since we disabled it ) As root user , fired the below command GRID_HOME/crs/install/rootcrs.sh -unlock GRID_HOME/crs/install/rootcrs.sh -lock Still after staring CRS , the gipcd was in unknown state
gipcd log after , applying the above workaroud still giving error
2021-02-10 17:19:35.164 : CSSCLNT:1286: clsssCommonClientExit: RPC failure, rc 3 2021-02-10 17:19:35.167 :GIPCDMON:1286: gipcdMonitorCssTerm: Successfully cleanup of connection to CSS 2021-02-10 17:19:35.167 :GIPCDMON:1286: gipcdMonitorThread: GIPCD received a shutdown msg from agent framework or some other thread died 2021-02-10 17:19:35.167 : GIPC:1286: gipcdsMemoryUnsubscribeMap: successfully un-subscribed the map file smid 00000000000018b8 tot 0 2021-02-10 17:19:35.168 : GIPCD:1286: gipcdSetThreadState: changing the status of monitorThread. current status gipcdThreadStatusOnline desired status gipcdThreadStatusOffline 2021-02-10 17:19:35.168 :GIPCDMON:1286: gipcdMonitorThread: Monitor thread is exiting.. 2021-02-10 17:19:35.176 : GIPCD:1: gipcdMain: All threads terminated 2021-02-10 17:19:35.176 : GIPCD:1: gipcdMain: GIPCD terminated
Tried this solution as well ora.gipcd not starting due to wrong ownership/permission (Doc ID 1949541.1) didn’t worked
Remove <ORACLE_BASE>/crsdata/<node>/output/gipcdOUT.trc and <ORACLE_BASE>/crsdata/<node>/output/gipcd.pid and restart.
Lastly , we tried removing and adding back the gipcd with proper permission with root user
Clusterware fail to start CRS-5809 and CRS-2680 after the roll back of a 12c OCW PSU (Doc ID 2051046.1)
[[ Login as root and execute the below commands ]]
1. verify if ohasd is running
[$ /oracle/app/12.2.0/grid/bin/crsctl check has
if not start only ohasd
# /oracle/app/12.2.0/grid/bin/crsctl start crs -noautostart
2. Delete the GIPCD resource.
# /oracle/app/12.2.0/grid/bin/crsctl delete res ora.gipcd -init -f
3. Delete the GIPCD resource type
# /oracle/app/12.2.0/grid/bin/crsctl delete type ora.gipc.type -init
4. Recreate the GIPCD resource type and the resource.
# cd /oracle/app/12.2.0/grid/crs/template
root@host(B)<-->{/oracle/app/12.2.0/grid/crs/template}:$ /oracle/app/12.2.0/grid/bin/crsctl add type ora.gipc.type -basetype ora.daemon.type -file /oracle/app/12.2.0/grid/crs/template/gipc.type -init
root@host(B)<-->{/oracle/app/12.2.0/grid/crs/template}:$ /oracle/app/12.2.0/grid/bin/crsctl add res ora.gipcd -type ora.gipc.type -attr "ACL='owner:root:rw-,pgrp:oinstall:rw-,other::r--,user:grid:rwx'" -init -unsupported
After performing the above steps try to shut down and start the CRS and look for gipcd status
root@host (B)<-->{/oracle/app/12.2.0/grid/bin}:$ ./crsctl stop crs -f
CRS-2791: Starting shutdown of Oracle High Availability Services-managed resources on 'host'
CRS-2673: Attempting to stop 'ora.mdnsd' on 'host'
CRS-2673: Attempting to stop 'ora.gpnpd' on 'host'
CRS-2673: Attempting to stop 'ora.evmd' on 'host'
CRS-2673: Attempting to stop 'ora.drivers.acfs' on 'host'
CRS-2677: Stop of 'ora.drivers.acfs' on 'host' succeeded
CRS-2677: Stop of 'ora.mdnsd' on 'host' succeeded
CRS-2677: Stop of 'ora.gpnpd' on 'host' succeeded
CRS-2677: Stop of 'ora.evmd' on 'host' succeeded
CRS-2793: Shutdown of Oracle High Availability Services-managed resources on 'host' has completed
CRS-4133: Oracle High Availability Services has been stopped.
Now start and verify if all services are up and running
root@host(B)<-->{/oracle/app/12.2.0/grid/bin}:$ ./crsctl start crs
CRS-4123: Oracle High Availability Services has been started.
root@host (B)<-->{/oracle/app/12.2.0/grid/bin}:$ ./crsctl stat res -t -init
Name Target State Server State details
Cluster Resources
ora.asm
1 ONLINE OFFLINE STABLE
ora.cluster_interconnect.haip
1 ONLINE OFFLINE STABLE
ora.crf
1 ONLINE OFFLINE STABLE
ora.crsd
1 ONLINE OFFLINE STABLE
ora.cssd
1 ONLINE OFFLINE host STARTING
ora.cssdmonitor
1 ONLINE ONLINE host STABLE
ora.ctssd
1 ONLINE OFFLINE STABLE
ora.diskmon
1 OFFLINE OFFLINE STABLE
ora.drivers.acfs
1 ONLINE ONLINE host STABLE
ora.evmd
1 ONLINE INTERMEDIATE host STABLE
ora.gipcd
1 ONLINE ONLINE host STABLE
ora.gpnpd
1 ONLINE ONLINE host STABLE
ora.mdnsd
1 ONLINE ONLINE host STABLE
ora.storage
1 ONLINE OFFLINE STABLE
292 total views, 5 views today