[Linux-cluster] Cluster Suite v3 IPMI fencing time

[Lon Hohberger]

On Sun, 2006-03-26 at 12:48 -0300, Celso K. Webber wrote:

> Mar 25 22:52:48 member1 cluquorumd[4112]: <warning> --> Commencing STONITH <--
> Mar 25 22:55:00 member1 cluquorumd[4112]: <notice> STONITH: member2 has been
> fenced!
> Mar 25 22:57:11 member1 cluquorumd[4112]: <notice> STONITH: member2 is no
> longer fenced off.
> 
> 
> Questions:
> 1. member1 does not take over the services until the fencing process
> completes, is that correct?

Yes.

> 2. if it is a correct behaviour that member1 waits for the fencing completion
> to take over services, how can I reduce the total fencing time? I didn't find
> any parameters under the cludb man page for this.

Turn off ACPI on both machines, or else IPMI will try a "graceful"
shutdown, which is exactly what you do *not* want.  If the machine is
dead or hung, it may not respond to the ACPI request at all.

Change the kernel command line to include "acpi=off" and chkconfig --del
acpid

Note that with IPMI, you should use the NIC with IPMI for only IPMI
traffic (or at least separate IPMI traffic from the cluster traffic), or
it can become a single point of failure.

-- Lon