17.12.2013 12:49, Michał Margula пишет:
Hello!
Hello!
Let me go trough it to see if I understand correctly:
W dniu 17.12.2013 08:57, Nikita Staroverov pisze:
primitive drbd_YOURVM ocf:linbit:drbd \
params drbd_resource="YOURVM" \
op monitor interval="29s" role="Master" \
op monitor interval="31s" role="Slave"
So here difference is target-role dropped? I had "Started" here.
Yes, there is difference. drdb primitive completely managed by
multistate resource. Target-Role in drbd primitive causes troubles.
I think this is the reason of your problem with resources actions looping.
primitive kvm_YOURVM ocf:heartbeat:VirtualDomain \
params config="/etc/libvirt/qemu/YOURVM.xml"
migration_transport="ssh" \
meta allow-migrate="false" target-role="Started"
is-managed="true" \
op start interval="0" timeout="120s" \
op stop interval="0" timeout="120s" \
op monitor interval="60s" timeout="120s"
Here seems to be the same (I use xen but it shouldn't matter).
Yes. It doesn't matter.
ms ms_drbd_YOURVM drbd_YOURVM \
meta master-max="1" master-node-max="1" clone-max="2"
clone-node-max="1" notify="true" target-role="Started" is-managed="true"
So here target-role is "Started" instead of "Master". And I have
master-max set to 2 and you have set to 1. It causes to have Master
(Primary DRBD) on one node and Slave (Secondary DRBD) on another?
Yes. If you want primary/primary, set master-max to 2. order promote
action changes ms drbd from secondary/secondary to primary/secondary(if
master-max=1) or to primary/primary(if master-max=2).
location kvm_YOURVM_prefer_YOURFIRSTSERVER kvm_YOURVM 50:
YOURFIRSTSERVER
It is optional? I mean it should not have any impact on DRBD
configuration.
Yes. I use it for preferred location of VM's.
order kvm_YOURVM_after_drbd_YOURVM inf: ms_drbd_YOURVM:promote
kvm_YOURVM:start
Same here.
No. Pacemaker with my configuration starts ms drbd in slave/slave
configuration. Due to ordering rule pacemaker promotes drbd before VM
starts.
In my configuration VM's works with primary/secondary drbd.
Why not dual primary? Is there a reason for that?
Thank you for your help!
I use KVM virtual machines with cache=writethrough, so i can't use live
migration.
_______________________________________________
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
