[ PCA ] After a 2.3.1 upgrade, a compute node is no longer able to start the o2cb service or mount any ocfs2 file systems (or the pool FS)
Last updated on SEPTEMBER 03, 2017
Applies to:Private Cloud Appliance - Version 2.1.1 and later
On Private Cloud Appliance, when upgrading compute nodes to release 2.3.1 of the PCA software, the compute node's upgrade fails with :
Tue Aug 29 14:12:48 2017 FATAL: Aborting upgrade job on server: ovcacn28r1. The upgrade is taking too long.
When logging on the compute node :
- The only ocfs2 file system mounted is the local repo with no heartbeat
- The node is out of the o2cb cluster
- The node exhibits a red cross icon in the Oracle VM Manager GUI
Additionally, when reviewing the logs, the compute node is unable to log on to 192.168.40.1 with no route to host
The bond for the 192.168.40.X network appears to be down, and finally the infiniband interfaces ib0 and ib1 do not exist anymore on the compute node.
Sign In with your My Oracle Support account
Don't have a My Oracle Support account? Click to get started
My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms