SG doesn't switch

[Ionut Grigorescu_2]

Hi,

I have still (I've opened before a thread - no solution) the folowing problem:
SG cluster on HPUX11.11, 2 nodes - one is the system server one database. When I'm powering-off the system server the packages running on it are startin successfully on the other node. When I'm disconnecting the LAN cables (hearthbeath + communication LAn - lan0 and lan1) the db server is rebooting because the nfs mounts are stale and is trying to start afterwards the cluster - but it's trying ever and aver to find the nfs server on the other node - unsuccesfully so it's remaining in this state and the cluster remains down the nodes are halted and the packages in unknown states. Enclosed the sgscusmx.cf file - the config for the cluster and the content of /etc/cmcluster/nfsserver directory - control scripts & logs - everything is the same on both nodes - any idea??

[Steven E. Protter]

Shalom,

There is a misconfiguration of something in the cluster.

I'd tail -f all the relavent log files and re-run the cluster failover test.

There is an accurate indicator somewhere.

If that doesn't work and its not too big a PITA, build the cluster again.

SEP

[melvyn burnard]

Please do not open a duplicate thread, reply to th eother thread:
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=1075019

It also may be a case of you shoul drather open a case with your local HP Response Centre

[Ionut Grigorescu_2]

melvyn: experience shown me that posting on old threads, didn't get any answers - so sorry, I need the answer, I always gave points so don't punish me for that.

anyway I cannot open a case against HP since the vendor is Nokia and the cluster was customized for Nokia and I have already a case against Nokia and they have no idea :-) and they are going to say - this is not supported, please make a request for feature to your Care Manager - do you think it's fair?