[Linux-cluster] fail to enable the vm in a cluster with vm service

[Chen, Ming Ming]

 Hi,

I have a two node cluster with a vm service. I can do  the vm migrate using clusvcadm (clusvcadm -M vm:vm297 -m node2). However if I use "clusvcadm -d vm:vm297" to disable/stop the vm, and then try to use "clusvcadm -e vm:vm297" to start it , then it will fail. However, I can manually  create the vm297  sing virsh (virsh create /abc/config/vm297.xml).  Any help and comment will be appreciated. Thanks in advance.

Ming



The following message is from the rgmanager.log file:

**** Stop the vm vm297 successfully *****

Apr 06 09:35:57 rgmanager 1 events processed

Apr 06 09:36:05 rgmanager Stopping service vm:vm297

Apr 06 09:36:05 rgmanager [vm] Using /abc/config//vm297.xml

Apr 06 09:36:26 rgmanager 1 events processed

Apr 06 09:36:39 rgmanager [script] Executing /etc/init.d/libvirtd status

****** start the vm 297 failed ********

Apr 06 09:36:49 rgmanager No other nodes have seen vm:vm297

Apr 06 09:36:49 rgmanager Starting disabled service vm:vm297

Apr 06 09:36:49 rgmanager [vm] Using /abc/config//vm297.xml

Apr 06 09:36:49 rgmanager [vm] /abc/config//vm297.xml is XML; using virsh

Apr 06 09:36:49 rgmanager [vm] virsh create /abc/config//vm297.xml

Apr 06 09:36:49 rgmanager start on vm "vm297" returned 1 (generic error)

Apr 06 09:36:49 rgmanager #68: Failed to start vm:vm297; return value: 1

Apr 06 09:36:49 rgmanager Stopping failed service vm:vm297

Apr 06 09:36:49 rgmanager Stopping service vm:vm297

Apr 06 09:36:49 rgmanager [vm] Virtual machine vm297 is

Apr 06 09:36:49 rgmanager Service vm:vm297 is recovering

Apr 06 09:36:49 rgmanager #71: Relocating failed service vm:vm297

Apr 06 09:36:49 rgmanager Service vm:vm297 is stopped

Apr 06 09:36:55 rgmanager 2 events processed





My cluster.conf file is :

<?xml version="1.0"?>

<cluster config_version="2" name="vmcluster">

      <logging debug="on"/>

      <cman expected_votes="1" two_node="1"/>

      <clusternodes>

            <clusternode name="shr297.cup.hp.com" nodeid="1">

                  <fence>

                        <method name="ilo">

                              <device action="reboot" name="shr298"/>

                        </method>

                  </fence>

            </clusternode>

            <clusternode name="shr330.cup.hp.com" nodeid="2">

                  <fence>

                        <method name="ilo">

                              <device action="reboot" name="shr331"/>

                        </method>

                  </fence>

            </clusternode>

      </clusternodes>

      <fencedevices>

            <fencedevice agent="fence_ilo" ipaddr="16.89.112.200" login="Administrator" name="shr298" passwd="lsdt4acsl"/>

            <fencedevice agent="fence_ilo" ipaddr="16.89.112.17" login="Administrator" name="shr331" passwd="lsdt4acsl"/>

      </fencedevices>

      <rm log_facility="local4" log_level="7">

            <service autostart="1" exclusive="0" name="storage" recovery="restart">

                  <script file="/etc/init.d/libvirtd" name="libvirt"/>

            </service>

            <vm autostart="1" exclusive="0" max_restarts="1" name="vm297" path="/abc/config/" recovery="restart" restart_expire_time="600"/>

      </rm>

</cluster>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/linux-cluster/attachments/20120406/b5bb7036/attachment.htm>