service guard process is getting killed

[RAJESH GANGADHARAN]

Hi,

We are having 2 single node cluster, where package will be inaccessible, and any cm commands like "cmviewcl" "cmhaltpkg" chmhaltcl, will not work, and following error message appears in the syslog.


ay 19 18:29:20 bahrain cmlvmd: Could not read messages from /usr/lbin/cmcld: Connection timed out
May 19 18:29:20 bahrain cmsrvassistd[6947]: Unable to communicate with ServiceGuard main daemon (cmcld): Network is unreachable
May 19 18:29:20 bahrain cmlvmd: Could not read messages from /usr/lbin/cmcld: Connection timed out
May 19 18:30:23 bahrain cmclconfd[1958]: The ServiceGuard daemon, /usr/lbin/cmcld[1960], died upon receiving signal number 6.

Can any one suggest why a process is getting termintaed by Signal 6???

I couldn't found the same in kill man page

Can any body help us to resolve this issue

Thanks in adv
Raj

[Mark Grant]

"kill -l" should confirm that signal 6 is SIGABRT. This suggests to me that it is being asked to die by something else but I could be wrong.

I would check the network and host resolution by doing pings and things.

[melvyn burnard]

you appear to be having network issues, and cmcld is hgetting SIGABRT, and is being aborted.
I would suggest you:
check all your networking
verify your Serviceguard is patched to latest patch

[Stephen Doud · Accepted solution]

I concur with Melvyn.

Also, review the ARPA transport patches on the server. It could be that an outdated version is due for replacement.

Also try cmcheckconf on the cluster configuration file - it may be out of date, or there may be a defective LAN on the server, or a defective network component.

A good diagnostic technique is to run a new cmquerycl, creating an output file, and comparing it against the configuration file that was used to build the cluster binary file (/etc/cmcluster/cmclconfig). If they don't match, then the hardware, network or LVM configuration has changed.

-StephenD

[Kent Ostby]

Not sure what your patching looks like, but there was a known problem for this that was originally fixed in PHNE_22507 which was an ARPA patch for 10.20.

What OS and version of SG are you running ?

[RAJESH GANGADHARAN]

Thanks for all updates

Here is some more information
Os version 11i
ServiceGuard A.11.15.00

output of cmqueycl is
Begin checking the nodes...
Looking for ServiceGuard nodes ... Done

Cluster Name Node Name Version Status
clusteruly1appoat
bahrain up to date

[Bharat Katkar]

Hi Raj,

Try doing "cmcheckcl" and see if it returns any error. It would be simple and better doing this first.

Regards,

[Bharat Katkar]

One more thing RAJ, it would better not to give more than 7 points if you are still expecting some repies because on the main menu it appears that your problem has been solved.

Regards,