On 8/2/07, Andre Heine <[EMAIL PROTECTED]> wrote:
> Hello all,
>
> I must update our NFS cluster with a new nfs-utils version.
> The new version doesn't need "nfslock" (it comes with the kernel), so
> I delete the entry with cibadmin...
>
> -----------------------
> cibadmin -D -X '<primitive class="lsb" id="nfslock_1" \
>                    provider="heartbeat" type="nfslock">'
> ----------------------
>
> I don't expect an failover, is that normal or can I control these
> behavior?
>
> Any hints?

i wouldn't expect a failover either... but its possible (since the
stickiness calculations will change)

to be sure, do this...

$ cibadmin -Ql > tmp.cib
$ vi tmp.cib    # remove nfslock
$ ptest -VVVV -X tmp.cib

that will tell you if anything will move
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to