So what causes clvmd to work this way in the first place? Why can't it play nice?
- Gary
"Alan A"
<[EMAIL PROTECTED]
m> To
Sent by: "linux clustering"
linux-cluster-bou <[email protected]>
[EMAIL PROTECTED] cc
Subject
11/19/2008 09:19 Re: [Linux-cluster] What is the
AM proper way to grow LVM/GFS volumes
Please respond to
linux clustering
<[EMAIL PROTECTED]
dhat.com>
This was great help - I rebooted the nodes one by one, and afterwards
lvextend command worked. I will try to reproduce the errors and see if
killing clvmd does the trick.
2008/11/19 Jeremy Lyon <[EMAIL PROTECTED]>
I have also encountered this problem several times, and although this
list seems to recommend running clvmd -R it has in fact never helped the
situation (I'm running centos 5.2). The only way I can solve this
problem is by rebooting all nodes in the cluster and then extending the
lv.
We have seen this too, but do not go the route of rebooting. The nice
thing about clvmd is that it's not required for the cluster to continue
running once up and established. It's purpose is to communicate LVM
metadata changes to all nodes in the cluster. So you can simply kill -9
the clvmd process on all nodes then run service clvmd start. This will
get clvmd back up and allow pv/vg/lv commmands to complete correctly.
-Jeremy
--
Linux-cluster mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/linux-cluster
--
Alan A.--
Linux-cluster mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/linux-cluster<<inline: graycol.gif>>
<<inline: pic31931.gif>>
<<inline: ecblank.gif>>
-- Linux-cluster mailing list [email protected] https://www.redhat.com/mailman/listinfo/linux-cluster
