vgHBPOCSHARED1 -wi-ao 20.00G -1 -1 253 21
R1gaUa-FDKx-1DEf-LT9d-vt86-o1Zz-l9LmuU
I now have a case raised with RedHat. I'll update if we make any progress.
Simon
-
Simon Hargrave szhargr...@ybs.co.uk
mailto:szhargr...@ybs.co.uk>
eid=10932>
Suggesting that reads of metadata were no always using O_DIRECT and doing
buffered reads. However, having applied this update, the symptoms persist.
I'll raise this as a support call.
-
Simon Hargrave szhargr...@ybs.co.uk
mailto:szhargr...@ybs.co.uk>
Technical Servic
Please read the warning at the end of this email
> As a rule of thumb, I always do a "vgscan" on all of the other nodes in the
> cluster, after creating a LV on a clustered VG.
> // Thomas
In this instance however, vgscan doesn't appear to f
Please read the warning at the end of this email
> Hi,
> Are you sure the clustered bit is set on the VG?
> http://sources.redhat.com/cluster/wiki/FAQ/CLVM#clvmd_clustered
> Bob Peterson
> Red Hat File Systems
Yes, the volume group was created wi
t clvmd,
service gfs2 stop
service clvmd stop
service clvmd start
service gfs2 start
Then everything returns to normal and the outputs match.
Is this something anyone has seen/can reproduce/has any idea about?
Versions are: -
lvm2-cluster-2.02.74-3.el5
lvm2-2.02.74-5.el5
Serve
