I’ve applied the PSU to our 12c cluster and stumbled across a few issues. This is a retrospective post on some of the problems I encountered. I went for the manual method and applied GI first then DB later on.

The opatch auto has an issue as can be seen in Martin’s Blog – Applying PSU in the lab environment There’s a good step by step guide on how to patch your cluster.

Not sure what has happened as everything seems to have gone well. I’ve been going through all the perl modules to try and work out what it is doing and where its gone wrong. Found out it uses checkpoints with the $GRID_HOME/bin/cluutil utility. I ran the checks manually and they returned the status expected. On a subsequent run the -postpatch seemed to go through and work ok until the mgmtdb database crashed

Rebooted node and fixed

I decided to plough on at this stage as it had successfully unlocked the grid home for patching. Oracle support also said to try and continue.

Same issue as previously experienced. Once again I manually run the $GRID_HOME/bin/cluutil utility but this time it gave me a few issues until it started to work.

Eventually I had a patched cluster all working well.

The database patch went in without an issue.

Update 19/05/2014

Please see my follow up post GI PSU for a better explanation of the issue and identifying the “root” cause

Leave a Reply

Your email address will not be published. Required fields are marked *

CAPTCHA * Time limit is exhausted. Please reload CAPTCHA.