ServiceGuard cluster locks

[Dolores Capers]

When I look in the syslog I get the following error:

WARNING: Cluster lock on disk /dev/dsk/c1t0d5 is missing!
until it is fixed, a single failure could cause all nodes in the cluster to crash.

I looked on the systems and the secondary was missing the special file for this disk. So, I used insf to create it and did the cmapplyconf again. But that didn't work. What am I missing?

[James R. Ferguson]

Hi:

To affect an update in the cluster lock disk, the cluster must be halted before you run 'cmapplyconf'.

...JRF...

[Rita C Workman]

I think this thread is pretty good, it gives some pretty good step by step instructions on this:

http://forums.itrc.hp.com/cm/QuestionAnswer/1,1150,0x6ceff841489fd4118fef0090279cd0f9,00.html

/rcw

[Carlos Fernandez Riera]

Perhaps you get a failed disk.

Check first:
dmesg
ioscan -fCdisk # search for
vi /var/adm/syslog/syslog.log
vgdisplay -v | grep stale
If this is the problem change failed disk. If the disk is hot plug you not need stop the cluster:

pvcreate
vgcfgrestore
vgchange -a y
vgsync

And cluster lock willbe regenerated.



If your problem is not a failed disk, you must stop cluster and reaply configuration.

There are more threads, and more complete, about this question.