Cannot online resource 'ResourceName' on node 'NodeName', because NSS on the node may not understand the newer NSS media associated

After applying the January 2016 Scheduled Maintenance Update, cluster resources may not come online. The following error is seen in /var/log/messages:

Feb 2 20:45:18 oec-bind04 kernel: [ 4882.058600] POOL_REVISION_CHECK (NCS_CRM_RES_T01:0:255): Cannot online resource 'ResourceName' on node 'NodeName', because NSS on the node may not understand the newer NSS media associated with the resource.

There is two-folded solution possible to this single problem.

1. Applying the March-2016-Hot-Patch

This issue is addressed with the March-2016-Hot-Patch

List patches by name on OES11sp2:

zypper pch | grep -i oes11sp2

Install the March 2016 Hot Patch

zypper up -t patch oes11sp2-March-2016-Hot-Patch

List patches by name on OES2015:

zypper pch | grep -i oes2015

Install the March 2016 Hot Patch

zypper up -t patch oes2015-March-2016-Hot-Patch

2. Check the cluster resource load script for variable substitution.

When pool/volume variable substitution is handled in a cluster resource load script, currently this is causing the same problematic behavior with respect the resource can not be brought online.

Even with the March-2016-Hot-Patch applied, the same error is observed.

Change :

exit_on_error nss /poolact=$VOLUME

to the actual pool name:

exit_on_error nss /poolact=<POOL_NAME>

With the January '16 Maintenance patch for OES11 SP2 and OES2015, the code to able to support new mixed node cluster functionality was introduced.

This check is based on the Pool Media version, but the code was not properly handling pools that were previously created on certain older versions of OES or NetWare, and which had undergone a series Pool Media version upgrades before