topic Re: Cluster fail over problem in Operating System - HP-UX

Cluster fail over problem

Kannandgl_1 — Fri, 07 Aug 2009 06:32:29 GMT

Dear Friends,

This is inform to my problem in fail over 2 node RAC cluster running on HP Ux.

1.yesterday we are changing the sharepoll & buffer cache value on the oracle data base side.. after changing the these value i getting fail over problems in the nodes.
2.Before When ever primary node fail it goes to automatic to 2 node.Now both nodes are running and up only things is fail over it will not te 2 node.
3.These value changes it will affect the cluster ? Please any help on this issue.

# cmviewcl -v

CLUSTER STATUS
cluster1 up

NODE STATUS STATE
hogisdata1 up running

Network_Parameters:
INTERFACE STATUS PATH NAME
PRIMARY up 0/3/1/0 lan0
PRIMARY up 0/4/2/0 lan2

PACKAGE STATUS STATE AUTO_RUN NODE
orapkg up running enabled hogisdata1

Policy_Parameters:
POLICY_NAME CONFIGURED_VALUE
Failover configured_node
Failback manual

Script_Parameters:
ITEM STATUS MAX_RESTARTS RESTARTS NAME
Service up 0 0 orapkg
Subnet up 10.38.1.0
Subnet up 10.10.10.0

Node_Switching_Parameters:
NODE_TYPE STATUS SWITCHING NAME
Primary up enabled hogisdata1 (current)
Alternate up enabled hogisdata2

NODE STATUS STATE
hogisdata2 up running

Network_Parameters:
INTERFACE STATUS PATH NAME
PRIMARY up 0/3/1/0 lan0
PRIMARY up 0/4/2/0 lan2

Regards
Kannan

Re: Cluster fail over problem

Lijeesh N G_1 — Fri, 07 Aug 2009 06:48:23 GMT

Hi Kannan,

Check in package control logs and syslog, i think you will get a clue.

Regards,
LIJEESH N G

Re: Cluster fail over problem

Vijaykumar_1 — Wed, 12 Aug 2009 03:15:30 GMT

Hi

Let me know on which node does the package runs now ?

Also grep for "NODE_NAME" in package configuration file and paste the o/p...

Re: Cluster fail over problem

Lijeesh N G_1 — Wed, 12 Aug 2009 03:21:25 GMT

Hi,

>>Let me know on which node does the package runs now ?
From the above output, It is clear that package"orapkg" is running on host "hogisdata1".

Regards,
LIJEESH N G

Re: Cluster fail over problem

Vijaykumar_1 — Wed, 12 Aug 2009 06:45:52 GMT

My apologies..I havent noticed that..

The change in values should not affect the cluster failover. You need to verify the configuration files for the package and also the logs..

Better grep for package name in the package control file, so that the particular logs can be extracted and analyzed.

In ServiceGuard its hard to troubleshoot without the error logs..

Re: Cluster fail over problem

Kannandgl_1 — Wed, 12 Aug 2009 08:07:51 GMT

Dear VijaiKumar,

Now the problem , resolved and its working fine... i am little bit confusion at the time..

Regards
Kannan