The process of posting to a mail list, thus documenting an issue, often
helps clarify the situation leading one to consider aspects that may not
have been explored ...this is one of those times.
I realised there was a 'debug' option in corosync.conf and, after changing
to 'on' then restarting corosync, discovered some interesting log
information like:
debug: native_assign_node: Could not allocate a node for res-ClusterMon
info: native_color: Resource res-ClusterMon cannot run anywhere
...which lead to the discovery that I needed some additional configuration:
location loc_res-ClusterMon_ol62dev res-ClusterMon 10: ol62dev
...but then spotted another issue:
debug: native_assign_node: Assigning ol62dev to res-ClusterMon
debug: custom_action: ol62dev res-ClusterMon_start_0 (cancelled :
quorum)
debug: RecurringOp: ol62dev res-ClusterMon_monitor_30000 (cancelled
: start un-runnable)
...uh huh ...maybe something to do with the global configuration which,
after comparing to another working system turned out to be:
no-quorum-policy="ignore"
...leaving me with a basic configuration that works:
crm(live)configure# show
node ol62dev
primitive res-ClusterMon ocf:heartbeat:ClusterMon \
params extra_options="-E /usr/local/bin/clusterevent.sh" \
operations $id="res-ClusterMon-operations" \
op monitor start-delay="0" interval="30" \
meta target-role="started"
location loc_res-ClusterMon_ol62dev res-ClusterMon 10: ol62dev
property $id="cib-bootstrap-options" \
expected-quorum-votes="2" \
stonith-enabled="false" \
no-quorum-policy="ignore" \
symmetric-cluster="false" \
dc-version="1.1.6-3.el6-a02c0f19a00c1eb2527ad38f146ebc0834814558" \
cluster-infrastructure="openais" \
last-lrm-refresh="1371781033"
...and a crm_mon process running with expected options
DB
_______________________________________________
Openais mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/openais
