Environment
Novell Open Enterprise Server 11 (OES 11) Linux Support Pack 2
SUSE Linux Enterprise Server 11 Service Pack 3 (SLES 11 SP3)
SUSE Linux Enterprise Server 11 Service Pack 3 (SLES 11 SP3)
Situation
After performing a rolling cluster upgrade (from OES 2 to OES 11 SP2), and all nodes were completely upgraded, Yast->Partitioner would not show any information for the local disk (on any node).
Upon checking with a number of tools like lvdisplay -a, vgs, and lvs; an error was returned (on every node in the cluster):
# /sbin/vgs
cluster request failed: Interrupted system call
Can't get lock for rootvg
Upon checking with a number of tools like lvdisplay -a, vgs, and lvs; an error was returned (on every node in the cluster):
# /sbin/vgs
cluster request failed: Interrupted system call
Can't get lock for rootvg
# /sbin/lvs
cluster request failed: Interrupted system call
Can't get lock for rootvg
Skipping volume group rootvg
# /sbin/lvdisplay -a
cluster request failed: Interrupted system call
Can't get lock for rootvg
Skipping volume group rootvg
cluster request failed: Interrupted system call
Can't get lock for rootvg
Skipping volume group rootvg
# /sbin/lvdisplay -a
cluster request failed: Interrupted system call
Can't get lock for rootvg
Skipping volume group rootvg
Resolution
In this case, a cluster restart 600 resolved the matter. As the command indicates, it takes all resources offline and restarts clustering services after 600 seconds (default is 60 seconds). If you have many active and large volumes, a longer time is helpful so they offline correctly.
Cause
One of the nodes did not release an lvm lock, or there was a corrupted NCS lock.