[LinuxFailSafe] Unsuccessful failover

[Christoph Biardzki]

Yes, I do have stonith (with APC MasterSwitch) and it works correctly WHEN
failsafe chooses to use it! But in this case it doesnt even try :(



On Fri, 3 May 2002, Lars Marowsky-Bree wrote:

> On 2002-05-03T11:13:43,
>    Christoph Biardzki <cbi@cebis.net> said:
>=20
> > I'm looking into a problem with failsafe - when I reboot the node (here
> > "pc92") with an NFS resource group, the other node tries to take over t=
he
> > resources - but if pc92 does not come up after the reboot (for any
> > reason) an "exclusivity test" fails. I understand it is necessary to te=
st
> > for exclusive resource allocation - but if the other node is down, the
> > "survivor" still cant do anything? Why isnt the "failed" (rebooted) nod=
e
> > simply reset?
>=20
> The other node is "simply" reset to ensure exclusivity if it drops out fr=
om
> the cluster; do you have STONITH / reset devices configured correctly?
>=20
>=20
> Sincerely,
>     Lars Marowsky-Br=E9e <lmb@suse.de>
>=20
> --=20
> Immortality is an adequate definition of high availability for me.
> =09--- Gregory F. Pfister
>=20
>=20