How to resolve the cluster upgrade state of [UPGRADE FINAL] after successfull...

刘泽宇

目标：
At the end of upgrading Grid Infrastructure (GI), the cluvfy performs the post upgrade check and reports the status.
Starting 18c or higher, the cluvfy does not change the cluster upgrade state from [UPGRADE FINAL] to [NORMAL] if the post upgrade check fails.
On the other hand, CRS, database, and all resources are running successfully on new upgraded version.

The reason for having the cluster upgrade state of [UPGRADE FINAL] is to alert users that the post upgrade check failed and there may be critical errors that should be fixed to ensure that the cluster remains stable.

Issue "crsctl query crs activeversion -f" to check the cluster upgrade state. The output should be
Oracle Clusterware active version on the cluster is [18.0.0.0.0]. The cluster upgrade state is [NORMAL]. The cluster active patch level is [0].

If the output of "crsctl query crs activeversion -f" shows "The cluster upgrade state is [UPGRADE FINAL]", then the post upgrade check by cluvfy failed and reported critical errors.


处理方法：
Fix the critical errors that cluvfy reported and rerun post upgrade check for cluvfy that failed earlier.

1. Issue "<NEW_UPRADED_GI_HOME>/bin/cluvfy stage -post crsinst -gi_upgrade -n all"
2. Fix the critical errors that above command reports
3. Rerun "<NEW_UPRADED_GI_HOME>/bin/cluvfy stage -post crsinst -collect cluster -gi_upgrade -n all"
4. Issue "<NEW_UPRADED_GI_HOME>/bin/crsctl query crs activeversion -f" and confirm that the cluster upgrade state is [NORMAL].
5. If the above command still reports that the the cluster upgrade state is [UPGRADE FINAL], repeat steps 1 to 3 and fix all critical errors.