cmviewcl can not view the cluster configurations

[mrmo07]

Hi boss,

getting the following message from from both node in cluster.
# cmviewcl
cmviewcl : Cannot view the cluster configuration. Either this node is not configured in a cluster, or else there is some obstacle to viewing the configuration. Check the syslog file for more information. For a list of possible causes, see the ServiceGuard manual for cmviewcl.

# uname -a
HP-UX dtcellm B.11.23 U ia64 4213628517 unlimited-user license

syslog(one node) is attached here.

please help.

[ani007]

Hi,
Please check whether cmcld is running or not."ps -ef|grep -i cmcld"
Regards,
Ani

[mrmo07]

Hi ani007,

what is the impact if cmcld is running or not. it that time let me check it.

Regards/Morshed.

[ani007]

it is cluster daemon.so you need to check whether it is running or not.Another thing i am not able to open ur attached file

[mrmo07]

hi please find the syslog as attached.

[ani007]

It's not opening.Please send me the op of "ps -ef|grep cmcld".

[Chandrahasa s]

Hi,

From syslog it looks like you have tried to start cluster and package on

Jul 9 05:12:57

********************************************
Jul 9 05:39:32 dtcellf cmcld: Service PKG*16897 terminated due to an exit(0).
Jul 9 05:39:32 dtcellf cmcld: Halted package unicorn_pkg on node dtcellf.
Jul 9 05:40:16 dtcellf sshd[8194]: Received disconnect from 172.16.8.124: 11: Terminating connection
Jul 9 05:41:09 dtcellf cmcld: (dtcellm) Started package unicorn_pkg on node dtcellm.
Jul 9 05:41:37 dtcellf cmcld: Request from node dtcellf to enable global switching for package unicorn_pkg.
Jul 9 05:41:37 dtcellf cmcld: Enabled switching for package unicorn_pkg.
Jul 9 05:46:18 dtcellf sshd[8707]: Did not receive identification string from 172.16.8.124
*********************************************
from above log the package is now up and running on node dtcellm.
this was happened on jul 9
there is no log entry related to cluster after that on syslog.

and there is log entry like
Did not receive identification string from 172.16.8.124

are you able to ping dtcellf??
are you able to ping dtcellm??

looks like some lan or host entry related issue on node dtcellf

Are you able to run cmviewcl from node dtcellm??

and also confirm below entry on file /etc/inetd.conf is not commented on both nodes.

auth stream tcp6 wait bin /usr/lbin/identd identd

if still problem persists,if you could able below logs we can light little more on your problem.

a)netstat -nvr from both nodes
b)cat cluster configuration file from both nodes.
c)syslog from both nodes.
d)/var/opt/resmod/log/event.log from both nodes.

Chandra

Assigning points will motivate each one in this forum.

[S. Ney]

I was able to open your second file.
Looks like you are attempting to start the cluster from cellf, cellf is your secondardary node. However you need to check your package and cluster config files as you have global switching disabled. So the package is stopping on cellm but not allowed to start on cellf.

Jul 9 05:13:01 dtcellf cmcld: 2 nodes have formed a new cluster, sequence #10
Jul 9 05:13:01 dtcellf cmcld: The new active cluster membership is: dtcellm(id=1), dtcellf(id=2)
Jul 9 05:13:01 dtcellf cmlvmd: Clvmd initialized successfully.
Jul 9 05:14:30 dtcellf cmcld: Request from node dtcellm to disable global switching for package unicorn_pkg.
Jul 9 05:15:17 dtcellf cmcld: (dtcellm) Halted package unicorn_pkg on node dtcellm.
Jul 9 05:15:44 dtcellf cmcld: Request from node dtcellm to start package unicorn_pkg on node dtcellf.
Jul 9 05:15:44 dtcellf cmcld: Executing '/etc/cmcluster/unicorn_pkg/unicorn_pkg.control start' for package unicorn_pkg, as service PKG*16897.
Jul 9 05:16:22 dtcellf cmcld: Service PKG*16897 terminated due to an exit(0).

I'd recommend to check your configuration files:
cmcheckconf -v -C /(ascii location)
cmcheckconf -v -P /(package location)

I'd also attempt to start one node:
cmrunnode -n dtcellm
cmrunpkg -n dtcellm unicorn_pkg
cmmodpkg -e unicorn_pkg
cmviewcl -v

If that works start the second node to see if it will join the cluster. If it does halt the package manually and run the package on dtcellf then cmmodpkg -e to allow for global switching.

[mrmo07]

# ps -ef |grep -i cmcld
root 2808 2807 0 Jul 9 ? 256:20 /usr/lbin/cmcld -j
root 20735 20338 1 11:09:34 pts/0 0:00 grep -i cmcld

[ani007]

I am not able to open the attachment...don't know why ..will it be possible to stop the cluster in onenode & before that open another session in putty & fire "tail -f /var/adm/syslog/syslog.log" .Pls send me the o/p of in notepad (attachment).

Please don't forget to assign points each of us..
Regards,
Ani

[mrmo07]

hi ..problem solved..i have just restarted inetd deamon with inetd -c.

thanks all.

Regards/Morshed.