<div dir="ltr"><div><div>Thank your for another idea, but I think I will pass ;)<br><br></div>I would like to use booth, but as I said, I also need location constraints based on an attribute-score (like ping). I don't think this is currently possible when using a multi-site-cluster.<br>
</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">2014-04-11 15:05 GMT+02:00 Lars Ellenberg <span dir="ltr"><<a href="mailto:lars.ellenberg@linbit.com" target="_blank">lars.ellenberg@linbit.com</a>></span>:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On Fri, Apr 11, 2014 at 10:02:59AM +0200, Christian Ciach wrote:<br>
> Thank you for pointing me to the environment variables. Unfortunately, none<br>
> of these work in this case. For example: Assume one node is currently the<br>
> master. Then, because of a network failure, this node loses quorum. Because<br>
> "no-quorum-policy" is set to "ignore", this node will keep being a master.<br>
> In this case there is no change of state, thus the notify-function of the<br>
> OCF-agent does not get called by pacemaker. I've already tried this, so I<br>
> am quite sure about that.<br>
<br>
<br>
</div>Very very hackish idea:<br>
<br>
set monitor interval of the Master role to T seconds<br>
and "fail" (+demote) if no quorum.<br>
<br>
(or use a dummy resource agent similar to the "ping" RA,<br>
and update some node attribute from there...<br>
then have a contraint for the Master role on that node attribute)<br>
<br>
in your promote action,<br>
refuse to promote if no quorum<br>
sleep 3*T (+ time to demote)<br>
only then actually promote.<br>
<br>
That way, you are "reasonably" sure that,<br>
before you actually promote,<br>
the former master had a chance to notice quorum loss and demote.<br>
<br>
But you really should look into booth, or proper fencing.<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
: Lars Ellenberg<br>
: LINBIT | Your Way to High Availability<br>
: DRBD/HA support and consulting <a href="http://www.linbit.com" target="_blank">http://www.linbit.com</a><br>
<br>
DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
_______________________________________________<br>
Pacemaker mailing list: <a href="mailto:Pacemaker@oss.clusterlabs.org">Pacemaker@oss.clusterlabs.org</a><br>
<a href="http://oss.clusterlabs.org/mailman/listinfo/pacemaker" target="_blank">http://oss.clusterlabs.org/mailman/listinfo/pacemaker</a><br>
<br>
Project Home: <a href="http://www.clusterlabs.org" target="_blank">http://www.clusterlabs.org</a><br>
Getting started: <a href="http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf" target="_blank">http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf</a><br>
Bugs: <a href="http://bugs.clusterlabs.org" target="_blank">http://bugs.clusterlabs.org</a><br>
</div></div></blockquote></div><br></div>