[Pacemaker] kvm VirtualDomain problem after reboot

[Ulrich Leodolter]

Hi,

i have setup a 2-node cluster and configured 4 virtualdomain
resources using iscsi storage.

when i reboot one machine (shutdown -r now) everything seems
to be ok,   virtual machines running on this node are migrated
away and continue to run on the other node.

but when the node comes up again i looks like it first checks
that none of the configured resources are running.

this checks fails VirtualDomain resources because libvirtd
is not running at this time, i can see a lot of errors like

Aug 17 15:30:06 pcmk1 lrmd: [1807]: info: RA output:
(p_virtdom_vm3:probe:stderr) Failed to connect socket to
'/var/run/libvirt/libvirt-sock': No such file or directory

the cluster does not startup does not succeed.
below you can see the cluster config.

what i am missing ?

thanks
ulrich



node pcmk1 \
	attributes standby="off"
node pcmk2 \
	attributes standby="off"
primitive p_iscsi lsb:iscsi \
	op monitor interval="30" \
	meta target-role="started"
primitive p_iscsid lsb:iscsid \
	op monitor interval="30" \
	meta target-role="started"
primitive p_libvirtd lsb:libvirtd \
	op monitor interval="30" \
	meta target-role="started"
primitive p_stonith_pcmk1 stonith:fence_ipmilan \
	params auth="password" ipaddr="192.168.120.171" passwd="secret" lanplus="true" login="Administrator" timeout="20s" power_wait="5s" verbose="true" pcmk_host_check="static-list" pcmk_host_list="pcmk1" \
	meta target-role="started"
primitive p_stonith_pcmk2 stonith:fence_ipmilan \
	params auth="password" ipaddr="192.168.120.172" passwd="secret" lanplus="true" login="Administrator" timeout="20s" power_wait="5s" verbose="true" pcmk_host_check="static-list" pcmk_host_list="pcmk2" \
	meta target-role="started"
primitive p_virtdom_vm1 ocf:heartbeat:VirtualDomain \
	params config="/etc/libvirt/qemu/vm1.xml" migration_transport="ssh" force_stop="false" \
	meta allow-migrate="true" is-managed="true" \
	op start interval="0" timeout="90s" \
	op stop interval="0" timeout="300s" \
	op migrate_to interval="0" timeout="120s" \
	op monitor interval="30s" timeout="90s" depth="0"
primitive p_virtdom_vm2 ocf:heartbeat:VirtualDomain \
	params config="/etc/libvirt/qemu/vm2.xml" migration_transport="ssh" force_stop="false" \
	meta allow-migrate="true" is-managed="true" \
	op start interval="0" timeout="90s" \
	op stop interval="0" timeout="300s" \
	op migrate_to interval="0" timeout="120s" \
	op monitor interval="30s" timeout="90s" depth="0"
primitive p_virtdom_vm3 ocf:heartbeat:VirtualDomain \
	params config="/etc/libvirt/qemu/vm3.xml" migration_transport="ssh" force_stop="false" \
	meta allow-migrate="true" target-role="Started" is-managed="true" \
	op start interval="0" timeout="90s" \
	op stop interval="0" timeout="300s" \
	op migrate_to interval="0" timeout="120s" \
	op monitor interval="30s" timeout="90s" depth="0"
primitive p_virtdom_vm4 ocf:heartbeat:VirtualDomain \
	params config="/etc/libvirt/qemu/vm4.xml" migration_transport="ssh" force_stop="false" \
	meta allow-migrate="true" target-role="Started" is-managed="true" \
	op start interval="0" timeout="90s" \
	op stop interval="0" timeout="300s" \
	op migrate_to interval="0" timeout="120s" \
	op monitor interval="30s" timeout="90s" depth="0"
group g_daemons p_iscsid p_iscsi p_libvirtd
clone cl_daemons g_daemons
location loc_p_stonith_pcmk1_pcmk1 p_stonith_pcmk1 -inf: pcmk1
location loc_p_stonith_pcmk2_pcmk2 p_stonith_pcmk2 -inf: pcmk2
colocation c_virtdom_vm1_on_daemons inf: p_virtdom_vm1 cl_daemons
colocation c_virtdom_vm2_on_daemons inf: p_virtdom_vm2 cl_daemons
colocation c_virtdom_vm3_on_daemons inf: p_virtdom_vm3 cl_daemons
colocation c_virtdom_vm4_on_daemons inf: p_virtdom_vm4 cl_daemons
order o_daemons_before_virtdom_vm1 inf: cl_daemons p_virtdom_vm1
order o_daemons_before_virtdom_vm2 inf: cl_daemons p_virtdom_vm2
order o_daemons_before_virtdom_vm3 inf: cl_daemons p_virtdom_vm3
order o_daemons_before_virtdom_vm4 inf: cl_daemons p_virtdom_vm4
property $id="cib-bootstrap-options" \
	expected-quorum-votes="2" \ 	
	dc-version="1.1.7-6.el6-148fccfd5985c5590cc601123c6c16e966b85d14" \
	no-quorum-policy="ignore" \
	cluster-infrastructure="openais" \
	last-lrm-refresh="1345208748"
rsc_defaults $id="rsc-options" \
	resource-stickiness="200"