Hi !
I am setting up my cluster and I have now some clustered file systems (gfs2)
under LVM.
Now I am doing some shutdown / reboot tests on divers node that are running
or not different services.
I noticed something at the verry begining of the startup :
Setting up Logical Volume Management: connect() failed on local socket: No
such file or directory
4 logical volume(s) in volume group VG now active
Skipping clustered volume group VGa
Skipping clustered volume group VGb
Skipping clustered volume group VGc
[FAILED]
So, I dug up in the rc.sysinit startup script and found at line 200 (on RHEL
6) this :
action $"Setting up Logical Volume Management:" /sbin/lvm vgchange -a
y --sysinit
That command (/sbin/lvm vgchange -a y --sysinit) return a non-zero value
when we have some clustered VG's, hence the fail in the startup.
That doesn't prevent any services from running, all is ok afterward, it's
purely esthetic...
Can other confirm this ? I did not find any bugs on this on the redhat bugs
site.
--
Linux-cluster mailing list
Linux-cluster@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/linux-cluster
