W dniu 23.12.2013 12:43, Nikita Staroverov pisze:
But it was causing trouble, trying to promote one of DRBDs and then
starting XEN. After it failed (because second DRBD was secondary) it
tried to stop both DRBDs and again. Also once it happened that XEN had
still device open and XEN was refusing to stop that VM.
So a mess, stonith, and so on.
Without pacemaker it works perfectly.
I haven't seen any colocation and order rules. So how should it work?
Sorry! Didn't paste enough. That is the rest:
<rsc_order first="clone-LVM" first-action="start"
id="DRBD-bilbo-root-after-LVM" score="INFINITY"
then="ms-DRBD-bilbo-root" then-action="promote"/>
<rsc_order first="clone-LVM" first-action="start"
id="DRBD-bilbo-squid-after-LVM" score="INFINITY"
then="ms-DRBD-bilbo-squid" then-action="promote"/>
<rsc_location id="XEN-bilbo-prefer-B" node="rivendell-B"
rsc="XEN-bilbo" score="50"/>
<rsc_colocation id="XEN-bilbo-with-DRBD" score="INFINITY">
<resource_set id="XEN-bilbo-with-DRBD-0">
<resource_ref id="XEN-bilbo"/>
</resource_set>
<resource_set id="XEN-bilbo-with-DRBD-1" role="Master"
sequential="false">
<resource_ref id="ms-DRBD-bilbo-root"/>
<resource_ref id="ms-DRBD-bilbo-squid"/>
</resource_set>
</rsc_colocation>
<rsc_order id="XEN-bilbo-after-DRBD" score="INFINITY">
<resource_set action="promote" id="XEN-bilbo-after-DRBD-0"
sequential="false">
<resource_ref id="ms-DRBD-bilbo-root"/>
<resource_ref id="ms-DRBD-bilbo-squid"/>
</resource_set>
<resource_set action="start" id="XEN-bilbo-after-DRBD-1">
<resource_ref id="XEN-bilbo"/>
</resource_set>
</rsc_order>
Sorry!
--
Michał Margula, [email protected], http://alchemyx.uznam.net.pl/
"W życiu piękne są tylko chwile" [Ryszard Riedel]
_______________________________________________
Pacemaker mailing list: [email protected]
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://bugs.clusterlabs.org
