[Pacemaker] Migrate vm on drbd in correct order?

[Matthias Teege]

Hi,

I'm try to setup a two node cluster for virtuell machines (KVM/libvirt)
with drbd as storage backend. For each vm I use two drbd devices. Starting
and stopping works but I cant migrate the VM to the second node. If i
switch a node to "standby" the VM stays "started" and there ist no call
to VirtualDomain.

I found the following in this list [0]

"Yes, migration only works without order-contraints the migrating service
depends on ... and no way to force it"

and I use order constraints because I need the drbd devices up before
the vm starts.

How do I migrate a vm or other ressources in the correct order? Is there
another way to migrate vm and drbd device, maybe start/stop?

My current config looks like this:

primitive vma_R ocf:linbit:drbd params drbd_resource="vma_R" \
  op monitor interval="29s" role="Master" \
  op monitor interval="31s" role="Slave" \
  op start timeout="240s" \
  op stop timeout="100s"

ms ms_vma_R vma_R \
  meta master-max="1" master-node-max="1" \
  clone-max="2" clone-node-max="1" \
  notify="true"

primitive vma_S ocf:linbit:drbd params drbd_resource="vma_S" \
  op monitor interval="29s" role="Master" \
  op monitor interval="31s" role="Slave" \
  op start timeout="240s" \
  op stop timeout="100s"

ms ms_vma_S vma_S \
  meta master-max="1" master-node-max="1" \
  clone-max="2" clone-node-max="1" \
  notify="true"

primitive kvm_vma ocf:heartbeat:VirtualDomain \
  params config="/etc/libvirt/qemu/vma.xml" \
  hypervisor="qemu:///system" migration_transport="ssh" \
  meta allow-migrate="true" \
  op start timeout="120s" \
  op stop timeout="180s" \
  op monitor depth="0" timeout="30" interval="10"

colocation vma_on_drbd \
  inf: ms_vma_R:Master ms_vma_S:Master kvm_vma:started
order vma_after_drbd \
  inf: ms_vma_S:promote ms_vma_R:promote kvm_vma

Thanks
Matthias

[0] http://www.mail-archive.com/pacemaker@oss.clusterlabs.org/msg10499.html