ODA X5 Server Patching To 12.1.2.10 Failed Due To Unable To Stop Clusterware On One Node

Asset ID:	1-72-2277272.1
Update Date:	2017-07-01
Keywords:

Solution Type Problem Resolution Sure

Solution 2277272.1 : ODA X5 Server Patching To 12.1.2.10 Failed Due To Unable To Stop Clusterware On One Node

Applies to:

Oracle Database Appliance - Version All Versions to All Versions [Release All Releases]
Information in this document applies to any platform.

Symptoms

Oracle Database Appliance X5-2

Failed to patch ODA X5 to 12.1.2.10.0 because failed to stop clusterware on one node during patching. The ODA was running on 12.1.2.7.0.

Also, run oaklci show disk returns not connect to oakd.

ERROR
-----------------------
> CRS-2673: Attempting to stop 'ora.crsd' on 'oda1'
> CRS-2790: Starting shutdown of Cluster Ready Services-managed resources on 'oda1'
> CRS-2679: Attempting to clean 'ora.ODATST.JMCTST1.inst' on 'oda1'
> CRS-2680: Clean of 'ora.ODATST.JMCTST1.inst' on 'oda1' failed
> CRS-2799: Failed to shut down resource 'ora.ODATST.JMCTST1.inst' on 'oda1'
> CRS-2794: Shutdown of Cluster Ready Services-managed resources on 'oda1' has failed
> CRS-2675: Stop of 'ora.crsd' on 'oda1' failed
> CRS-4000: Command Stop failed, or completed with errors. ,

STEPS
-----------------------
[root@oda1 ~]# oakcli show version
Version
-------
12.1.2.10.0

[root@oda2 ~]# oakcli show version
Version
-------
12.1.2.7.0

Unable to complete ODA patching.

Changes

Attempting to patch ODA Server after manually deleting a test database outside of the OAKCLI and Oracle Clusterware/GI.

Cause

Issues stopping the Cluster database resources:

  ora.ODATST.db

  ora.ODATST.JMCTST1.inst

  ora.ODATST.JMCTST2.inst

Determined these database were long since deleted, but not removed from the Clusterware/OCR.

WebEx session with customer.

Solution

Determined the Clusterware/G was not starting ODA node oda1 due to clusterware being disabled.

Correct by issuing:

  crsctl enable crs

Now able to start the Clusterware on ODA node oda1.

Regarding the ODA Server patch failing.

Found issues stopping the following resources:

  ora.ODATST.db

  ora.ODATST.JMCTST1.inst

  ora.ODATST.JMCTST2.inst

Determined these database were long since deleted, but not removed from the Clusterware/OCR.

Manually removed using:

  crsctl delete res ora.ODATST.db -unsupported

  crsctl delete res ora.ODATST.JMCTST1.inst -unsupported

  crsctl delete res ora.ODATST.JMCTST2.inst -unsupported

Tried re-running the ODA Server patching using:

  oakcli update –patch 12.1.2.10 --server -local

This returns error:

Verified customer was able to successfully complete Server patching of Infra and GI components using command, 'oakcli update –patch 12.1.2.10 --server -local'. Corresponding ODA components on oda1 node show 'Up-to-Date'.

Customer proceeding to patch Server on second ODA node, oda2.

Attachments

This solution has no attachment