hi, i'm hoping somebody can help me get my STONITH device working
properly. i'll try to be brief... :)
two node cluster (names beast and mystique, respectively), each node
has a DRAC5 card. i've elected to use the external/ipmi STONITH
agent. testing my configuration with the following STONITH CLI command
reboots beast perfectly:
mystique$~: stonith -t external/ipmi -p "beast beast_drac hacluster
supersecret" -T reset beast
however, i can't seem to get heartbeat to automagically STONITH beast
when appropriate. for my test, i simply 'killall -9 heartbeat' on
beast, and wait. my resources fail over just fine to mystique, but no
STONITH for the beast :(
i've tried both version 1 and version 2 configurations to no avail.
i'm attaching those here for reference.
VERSION 1:
// STONITH line in ha.cf
mystique$~: cat /etc/ha.d/ha.cf | grep ^stonith
stonith external/ipmi /etc/ha.d/drac5.cfg
// corresponding drac5.cfg file
mystique$~: cat /etc/ha.d/drac5.cfg
beast beast_drac hacluster supersecret
// perms on drac5.cfg
mystique$~: ls -al /etc/ha.d/drac5.cfg
-rw------- 1 root root 55 Aug 30 21:48 /etc/ha.d/drac5.cfg
VERSION 2:
// relevant resource
<primitive id="resource_stonith_beast" class="stonith"
type="external/ipmi" provider="heartbeat">
<meta_attributes id="resource_stonith_beast_meta_attrs">
<attributes>
<nvpair id="resource_stonith_beast_metaattr_target_role"
name="target_role" value="started"/>
</attributes>
</meta_attributes>
<instance_attributes id="resource_stonith_beast_instance_attrs">
<attributes>
<nvpair id="9cb48aa3-82c0-43df-b9a9-06cad731a67e"
name="hostname" value="beast"/>
<nvpair id="a5d8d0b8-f753-47ef-9381-736ed76822d0"
name="ipaddr" value="beast_drac"/>
<nvpair id="f3bb9451-003a-49b6-b12b-89cdff1a40a7"
name="userid" value="hacluster"/>
<nvpair id="f1954c51-f7af-474a-b63b-21aa22921c47"
name="passwd" value="supersecret"/>
</attributes>
</instance_attributes>
<operations>
<op id="223c6d56-f8ec-4f42-85ca-214775f0fc2c" name="monitor"
interval="15" timeout="15" start_delay="15" prereq="nothing"/>
<op id="f0f6e6fd-211b-4bf8-9bbc-3f1fabf7fcba" name="start"
timeout="15" prereq="nothing"/>
</operations>
</primitive>
// force the resource to live on the right node
<rsc_location id="location_stonith_beast_on_mystique"
rsc="resource_stonith_beast">
<rule id="prefered_location_stonith_beast_on_mystique"
score="INFINITY">
<expression attribute="#uname" id="4fbbd7c1-4fc2-44ee-
b9d4-1deccd505728" operation="eq" value="mystique"/>
</rule>
</rsc_location>
i'm finding these entries in ha-log, which seem to indicate that the
daemon is running:
tengine[6803]: 2008/08/30_22:02:47 info: te_connect_stonith:
Attempting connection to fencing daemon...
tengine[6803]: 2008/08/30_22:02:48 info: te_connect_stonith: Connected
at this point i'm at a loss, so any help/tips/advice would be greatly
appreciated.
thanks!
chad
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
