[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Linux-cluster] Child vs Resource




Yes, unless you specify __independent_subtree for a resource, then it
and its children are independent and attempted to be restarted as a
separate operation from a full service restart.  Only if that partial
restart fails in this case does a full service restart occur.

I'm not seeing this behavior nor what is documented at http://sources.redhat.com/cluster/wiki/ResourceTrees in the Failure Recovery & Independent Subtrees.  Is __independent_subtree not applicable with RHEL 4.6 RHCS?  Regardless of the configs I've used the entire service gets restarted instead of just the specific failed resource

rgmanager-1.9.72-1


<?xml version="1.0"?>
<cluster alias="cluster" config_version="25" name="cluster">
        <fence_daemon post_fail_delay="0" post_join_delay="3"/>
        <clusternodes>
                <clusternode name="hostA" votes="1">
                        <multicast addr="239.192.127.204" interface="bond0"/>
                        <fence>
                                <method name="1">
                                        <device blade="2" name="blade_fence"/>
                                </method>
                        </fence>
                </clusternode>
                <clusternode name="hostB" votes="1">
                        <multicast addr="239.192.127.204" interface="bond0"/>
                        <fence>
                                <method name="1">
                                        <device blade="3" name="blade_fence"/>
                                </method>
                        </fence>
                </clusternode>
        </clusternodes>
        <cman expected_votes="1" two_node="1">
                <multicast addr="239.192.127.204"/>
        </cman>
        <fencedevices>
                <fencedevice agent="fence_bladecenter" ipaddr="192.168.2.16" login="rhcs_fence" name="blade_fence" passwd="XXXXXXX"/>
        </fencedevices>
        <rm>
                <failoverdomains>
                        <failoverdomain name="allnodes" ordered="0" restricted="0">
                                <failoverdomainnode name="hostA" priority="1"/>
                                <failoverdomainnode name="hostB" priority="1"/>
                        </failoverdomain>
                </failoverdomains>
                <resources>
                        <ip address="192.168.3.64" monitor_link="1"/>
                        <clusterfs device="/dev/vgclgfs/testlv" force_unmount="0" fsid="10517" fstype="gfs" mountpoint="/gfs/testlv" name="testlv" options=""/>
                        <script file="/gfs/testlv/cluster-scripts/rc.oracle" name="Oracle_test"/>
                        <ip address="192.168.3.65" monitor_link="1"/>
                        <ip address="192.168.3.66" monitor_link="1"/>
                </resources>
                <service autostart="1" domain="allnodes" name="oratest">
                        <clusterfs ref="testlv"/>
                        <ip ref="192.168.3.64"/>
                        <script file="/gfs/testlv/cluster-scripts/jeremy/rhcs.oracle.db" name="oratest_db" __independent_subtree="1">
                                <script file="/gfs/testlv/cluster-scripts/jeremy/rhcs.oracle.listener" name="oratest_list"/>
                        </script>
                </service>
                <service autostart="1" domain="allnodes" name="tnwfst1">
                        <ip ref="192.168.3.65"/>
                </service>
                <service autostart="1" domain="allnodes" name="tnwfst2">
                        <ip ref="192.168.3.66"/>
                </service>
        </rm>
</cluster>
[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]