Need to configure autorestart

[lahenrich]

Hi

We are trying to configure VCS on our systems.  We have the groups set up and everything works well in "sunny" day situations, meaning we can offline, online and switch the VCS groups between the servers.  What doesn't work is while testing, we manually shut down an oracle database and/or listener to simulate a disaster, and VCS makes no attempt to restart it.  hagrp -state does show the group as "partial" and then "online" when we manually bring it back up.  But we need VCS to handle this automatically.

What do I need to configure to get this to work?

Thanks,
Laura

[mikebounds · Accepted solution]

Issue could be:

1. IntentionalOffline attribute of the Oracle type is set to 1, so VCS knows you are shutting Oracle down.  To check run "hatype -display Oracle  | grep -i intention" and if IntentionalOffline is set to 1, then I would recommend killing processes, instead of shutting Oracle down cleanly foryour test or set IntentionalOffline to 0
2. Critical attribute for Oracle resources are set to 0.  To check run "hares -display | grep Critical" to see if Oracle resources have Critical of 0.  If they are set to 1 using "hares -modify res_name Critical 1"

Mike

[lahenrich]

Hi Mike

Thanks for your quick response!  Your first solution did not apply since IntentionalOffline was already set to 0.  But your second solution did the trick.  Both my database group and listener group had Critical set to 0.  I reset them to 1, retested and VCS brought them back up.

Thanks!

Laura