need to know the VCS resource status if intentional recycle occur for DB

[sureshpeters]

Could you please clarify the below scenerio:-

Our DB admin required to recycle the DB instance from Oracle level (without VCS knowledge). the critical of this resource and dependent of all resources of this is 0. 

Will the resource go OFFLINE/Online (or) FAULTED?

And the same scenerio applied on filesystem resource lets say , unmounting and mounting the file system resource manually by unmount command (without VCS knowledge) where critical is set 0 including all the dependent resource criticality is 0.

Will the resource go OFFLINE/ONLINE (or) FAULTED?

Please provide complete details of how this works.

Thanks

[Daniel_Matheus]

Hi Suresh,

If a resource is configured as non-critical (by setting the Critical attribute to 0) and no resources depending on the failed resource are critical, the service group will not fail over. VCS takes the failed resource offline and updates the group status to online|partial. The attribute also determines whether a service group tries to come online on another node if, during the group's online process, a resource fails to come online.

If you are agent is at least V51 or later you could also set the intentionaloffline attribute to 1.

https://sort.symantec.com/public/documents/sf/5.0MP3/linux/html/vcs_agent_dev/ch_static_attributes21.html#1138380

https://sfdoccentral.symantec.com/sf/5.1/aix/html/vcs_notes/ch01s08s19.htm

If your Oracle agent is older than V51 I would suggest that you freeze the service group before you shutdown Oracle or even better write a script that the Oracle admin can execute which will stop Oracle using VCS commands, this way no problem at all is to be expected.

You could then make it available using sudo for example.

Hope this helps,

Dan

[mikebounds]

VCS will fault the resource and call the clean script for that resource, so if a DBA has brought the database part the way down, VCS will bring it all the way down if VCS sees it as not fully online.  All dependent resources will be offlined.  If the resource that faulted or any of the dependent resources that were offlined were critical, then the service group will failover, otherwise the service group will stay where it is.

As Dan say, I would recommend freezing the servicegroup before DB is recycled.  You can allow a DB to freeze service group by creating an "Oracle" VCS user and giving it "Operation" privilige on the Oracle service group (technically this also gives them permisson to offlined ANY resource in the Oracle service group, like Mount DG and IP).

Mike

[Alok_Sontakke]

It depends on whether the Oracle agent can detect it as intentional ofline. Please check the IntentionalOffline attribute of the Oracle type to verify that it supports IntentionalOffline.

If IntentionalOffline is 1, then the Oracle agent will report an intentional offline and VCS will mark the resource as OFFLINE (not FAULTED). Further actions depends on the value of attribute ExternalStateChange.

If IntentionalOffline is 0, then Oracle agent will report offline and VCS will mark the resource as FAULTED. Further actions depends on the value of Critical attribute.

The Mount resource does not support IntentionalOffline, hence on unmount, VCS will mark the resource as FAULTED and take action according to the Critical attribute.

Alok

[sureshpeters]

So, Does it mean Even if the resource (DB or Mount) Critical attribute is 0 and all dependent resources were also Critical attribute 0. if the DB and Mount resource are recycled (without VCS knowledge) then the resource will be Faulted and service group status turn to FAULTED. Is this correct?

[mikebounds · Accepted solution]

No resource would fault but service group would not fault (so service group would not fail to other servers) so resource state will be faulted and service group state will be Partial.

Mike

[sureshpeters]

Thanks

[sureshpeters]

I would like to remove the SG, will the below steps work? hagrp -offline SG –sys server1 hares –delete resources - delete parent resources first since resources are dependent on others Parent - tony, Child - ting So, run command hares -delete tony (first) hagrp –delete SG