HACMP V5.1 Cluster in unstable state

[Rizwan Abbasi]

Hi All,
May need the group wisdom to solve this issue....

I am trying to install two node rotoating cluster over two p630s and
two SSAs RAID5 D40 disks with two VGs under one rotating resource
group, under the AIX OS 5.3 ML3 and HACMP V5.1 with latest fixes
applied from the WEB and the RSCT version 2.3.3.3, i am using TMSSA
for non IPs network and using ip aliasing.
The problem I am facing is that after succesful varification and
synchronizing of cluster when I start the cluster its shows cluster in
unstable state under cluster problem determination and view cluster
status menu with all the intertface in down state and resource group
showing as acquiring. The event that shows not completed in hacmp.out
is node_up.rc.

I am using following IPs

10.24.5.222 appsprod1_boot1
10.25.5.222 appsprod1_boot2
10.24.5.223 appsprod2_boot1
10.25.5.223 appsprod2_boot2
10.4.5.210 appsprod #service ip lable

Any help or hints or tips will be of most value to me. Thanks in
advance for the help.

[Pete's]

abb...@hotmail.com (Rizwan Abbasi) wrote in message news:<a1516653.04080...@posting.google.com>...

First, I recommend going to 5.1.0.5 on you cluster software. Second,
is the node halting after startup? Are you using Oracle(Non-RAC) in
this cluster? If so, make sure you have renamed the following files
back to their original names(in /lib):
libha_gs.a
libha_gs64.a
libha_gs64_r.a
libha_gs_r.a

HTH,
Pete's

[Riz]

Hi,

Thanks for the reply, I am using the latest fixes i.e. HACMP 5.1.0.5, also
here is the latest update....
I find out there was some problem with the tmssa link so I removed the
tmssa network from HACMP and am now using the rs232 native s3 on p630s for
HBs, now the symptoms are changed a bit. The nodes comes up normally and
shows normal under the clstat, however when I try to failoever to the backup
node the resource group shows acquiring in view cluster state. Also the
serivice IP move on the standby node but the VGs do not move compeletly
however it do shows in lsvg -o but when I do the lsvg -l datavg, the command
hangs. The cluster state shows unstable in clstat command. any tips pls
pass on to me.

Kind Regards

Riz

"Pete's" <empet...@yahoo.com> wrote in message
news:6724a51f.04081...@posting.google.com...