cmcluster

[Alex Lavrov]

Hello,
I have two L Class servers in the cluster, this night one of them lost connection to the other and rebooted. after it came up, the cluster refuses to work on this computer.
There is connection between them telnet, ftp, ping works fine), all the cluster configurations remained the same, but I get these messages:

[34] serv1 # /sbin/init.d/cmcluster start
vxvm:vxdisk: ERROR: IPC failure: Configuration daemon is not accessible

cmrunnode: Unable to determine the nodes on the current cluster
cmrunnode: Either no cluster configuration file exists, or the file is corrupted, or cmclconfd is unable to run
Local node is not currently configured in cluster
ERROR: Unable to join cluster

Configuration file exists ...
and the messages are from cmclconfd, i guess ...

Any suggestions?

thnx

[Tim Adamson_1]

Sounds like the problem lies with VxVM.

Does the cluster use VxVM disks?

Have you checked that VxVM is running properly and the necessary VxVM disk groups are available?


Tim.

[Zigor Buruaga]

Hi,
Check this link, it mention the patch PHSS_24536
http://archives.neohapsis.com/archives/hp/2002-q1/0094.html

Hope this helps.
Kind regards,
Zigor

[Alex Lavrov]

thanx for the help, but I found the problem.
someone removed the names of the other servers in the cluster from .rhosts on this one.

[Geoff Wild]

Just a tip - instead of .rhosts, use cmclnodelist in /etc/cmcluster

Format like:
system3 root
system3 oper1
system4 root
system4 oper1



That allows root on both boxes as well, you can add other users that you may want to allow to run cluster commands.



Rgds....Geoff

[Jeff Schussele]

Hi Alex,

This sounds like a classic heartbeat problem.
What subnet holds the heartbeats?
Check them because if you lose ALL heartbearts, the system(s) will TOC - read REBOOT.
IF they're on a "public" LAN, they shouldn't. They should be on a private LAN - just to cover these situations.
SO - check the heartbeat LAN & if trouble - correct & all should be well again.

HTH,
Jeff