No subject

terminate other nodes connection to the SAN via the SAN switch or the
SAN controller itself.  This can be used as a primary fencing
mechanism and will prevent data loss in the contexts you mention.

On 7 January 2011 08:38, Robert van Leeuwen <vanleeuwen at stone-it.com> wrote=
:
>> In a simple case, we want to prevent two nodes using a shared file-syste=
m
>> without coordination. =A0If they loose contact with each other and can't
>> coordinate their access, one must die. =A0(Even if you make it hard for =
them to
>> loose contact by providing multiple communication channels, you have to =
assume
>> it will happen and, so, we work with this case.)
>
> Not a real help but...
> My preferred way is to create a 3 node cluster even if one node is only f=
or voting and will not run any resources.
> With 3 nodes you can correctly "vote" about who's online & offline (2 vs =
1 vote).
> When none of the 3 nodes see each others quorum will be lost on all nodes=
 and the cluster will freeze (or whatever you configure).
>
> Some cluster suites use a quorum disk for the creating the 3rd vote but a=
s far as I know Pacemaker does not support this.
>
> I'm afraid if a third node is not possible you will have to make absolute=
ly sure the network link stay's up but it won't be 100% fool proof if someb=
ody starts messing with the cables...
> (you're best bet will be multi-path + cross cable e.g.)
>
>
> Robert
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=3DP=
acemaker
>



--=20
Best Regards,

Brett Delle Grazie