Hi!
I've two cluster-nodes, both running pingd (as a clone), to keep ressources
from starting on nodes which have not obvious connection to the network. The
ping-nodes are:
- appl01 (10.10.10.202)
- appl02 (10.10.10.203)
- Default GW (10.10.10.254)
Before shutting down the interfaces on node2 is see for "cibadmin -Q | grep
status | grep ping":
<nvpair name="pingd" id="status-cms-appl01-pingd" value="3000"/>
<nvpair name="pingd" id="status-cms-appl02-pingd" value="3000"/>
The command crm_mon shows, that the cluster is fully operational; rcdrbd status
shows a connected redundant blockdevice with primary and secondary; all fine
here.
When I now shut down the interfaces (ifconfig eth0 down ; ifconfig eth1 down)
of appl02, the following is the output of cibadmin -Q ...:
<nvpair name="pingd" id="status-cms-appl01-pingd" value="3000"/>
<nvpair name="pingd" id="status-cms-appl02-pingd" value="3000"/>
Yes! Nothing has changed! Appl02* sees itself as fully connected and starts the
clustered services (takes over drbd, the common IP etc.). When I ping my nodes
manually, the node only reaches itself. According to my understanding of the
matching cib-entry:
primitive pingy_res ocf:pacemaker:ping \
params dampen="5s" multiplier="1000" host_list="10.10.10.202
10.10.10.203 10.10.10.254" \
op monitor interval="60s" timeout="60s" \
op start interval="0" timeout="60s"
clone pingy_clone pingy_res \
meta target-role="Started"
location only-if-connected appl_grp \
rule $id="only-if-connected-rule" -inf: not_defined pingd or pingd lt
2000
...the value for pingd should be 1000 now, not 3000 and thus the ressources in
the appl_grp should never be started on this system. Well... they do.
Interestingly, the logs show, that pingd notices that node1 and the default-gw
are not reachable anymore. I read:
Ping [15560]: ERROR: Unexpected result for "ping -n -q -W 3 -C 5 10.10.10.202
2: connect: Network is unreachable".
When I now reconnect my interfaces, the cluster has run into a split brain,
which I can resolve by rebooting both nodes. That's not too desirable. :(
Any ideas?
Thanks for your help,
Andreas Stallmann
*strangely, appl01 notices, that one of the three ping-nodes is missing and
shows:
<nvpair name="pingd" id="status-cms-appl01-pingd" value="2000"/>
<nvpair name="pingd" id="status-cms-appl02-pingd" value="3000"/>
How come it doesn't change the value of appl02 by the way? As the cluster is
disconnected, appl01 should have no knowledge of appl02 ping status.
------------------------
CONET Solutions GmbH, Theodor-Heuss-Allee 19, 53773 Hennef.
Registergericht/Registration Court: Amtsgericht Siegburg (HRB Nr. 9136)
Gesch?ftsf?hrer/Managing Directors: J?rgen Zender (Sprecher/Chairman), Anke
H?fer
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
