[Linux-cluster] info on modclusterd in cluster 3

Gianluca Cecchi gianluca.cecchi at gmail.com
Wed Aug 5 09:20:06 UTC 2009


Hello,
I have a cluster 3 on 2 x F11 x86_64 nodes.
main components are:
fence-agents-3.0.0-14.fc11.x86_64
clusterlib-3.0.0-20.fc11.x86_64
resource-agents-3.0.0-12.fc11.noarch
lvm2-cluster-2.02.48-2.fc11.x86_64
modcluster-0.16.1-1.fc11.x86_64
corosynclib-1.0.0-1.fc11.x86_64
cman-3.0.0-20.fc11.x86_64
corosync-1.0.0-1.fc11.x86_64
rgmanager-3.0.0-20.fc11.x86_64

It seems that on both nodes modclusterd is not able to start
with this log in /var/log/clumond.log
Mon Aug  3 17:04:31 2009 clumond: started
Mon Aug  3 17:04:31 2009 clumond: unhandled exception in main():
ServerSocket(port=16851): socket() failed: Address family not
supported by protocol
Mon Aug  3 17:04:31 2009 clumond: died

If I try to manual restart it (with cman, rgmanager, ricci aleady
started and up), I get:
[root ~]# service modclusterd start
Starting Cluster Module - cluster monitor: Setting verbosity level to LogBasic
                                                           [FAILED]

Is this service supposed to start automatically (aka chkconfiged on)
in cluster 3?

In rh 5.3 my chkconfig situation is this:

cman           	0:off	1:off	2:on	3:on	4:on	5:on	6:off
modclusterd    	0:off	1:off	2:on	3:on	4:on	5:on	6:off
openais        	0:off	1:off	2:off	3:off	4:off	5:off	6:off
qdiskd         	0:off	1:off	2:on	3:on	4:on	5:on	6:off
rgmanager      	0:off	1:off	2:on	3:on	4:on	5:on	6:off
ricci          	0:off	1:off	2:on	3:on	4:on	5:on	6:off

and when I have it:
clvmd          	0:off	1:off	2:on	3:on	4:on	5:on	6:off

So that when I have to do maintenance on a node, leaving it open but
without cluster infrastructure active I do :
1) eventual relocation of active services
2) stop of related services:
service modclusterd stop
service rgmanager stop
service ricci stop
service clvmd stop (if present)
service qdiskd stop
service cman stop

At this moment in my F11 nodes I have the same chkconfig setup for
these services (apart from qdisk that I have not configured at the
moment).

Has been changed anything in startup scripts config setup?
Also,
Description for modclusterd init script:
# description:  Starts and stops Ricci Cluster Module - cluster monitor

Description for ricci init script:
# description:  Starts and stops Red Hat Cluster and Storage Remote \
#               Configuration Interface (ricci)

Not very clear (at least to me) what are the relative functions of
these two services.

Thanks,
Gianluca




More information about the Linux-cluster mailing list