Video Screencast Help

advice for DiskGroup and Volume resources for SFCFSHA environment

Created: 04 Apr 2013 • Updated: 07 Apr 2013 | 12 comments
Zahid.Haseeb's picture
This issue has been solved. See solution.

Environment

Cluster Nodes = Two

OS = rhel 6.2

sfcfsha = 6.0

Default resources of cvm Service Group under SFCFSHA Cluster = vxfsckd, cvm_clus, cvm_vxconfigd

 

I want to create DIskGrop and Volume resource. I am trying to create CFSmount and CVMVolDg. which I feel correct. But both resources are not getting online and getting faulted. Advice please.

 

 

Operating Systems:
Discussion Filed Under:

Comments 12 CommentsJump to latest comment

mikebounds's picture

What have you done so far and what errors are you getting.  One thing to make sure is that your diskgroup is imported and shared as CVMVolDg activates diskgroup - it does not import it.  So if your diskgroup was created with "-s" flag, then deport diskgroup and import with -s flag.

Mike

UK Symantec Consultant in VCS, GCO, SF, VVR, VxAT on Solaris, AIX, HP-ux, Linux & Windows

If this post has answered your question then please click on "Mark as solution" link below

Zahid.Haseeb's picture

- install sfcfsha version 6

- configure sfcfsha

- Diskgroup created via vxdiskadm

- Now what resource should I add now  ?

Any comment will be appreciated. Mark as Solution if your query is resolved
__________________
Thanks in Advance
Zahid Haseeb

zahidhaseeb.wordpress.com

Zahid.Haseeb's picture
vxdg -s init DG DiskName
cfsdgadm add DG all=sw

 

vxassist -g DG make VOL XXg

cfsmntadm add DG Vol /mnt all=cluster

======================================================

After doing above one Service Group Created with two resources . One is resource is for DiskGroup and second is for Mount...

I think I did it but one question. Cant we create the DiskGroup and Volume resource manually from Java Console instead of as created automatically from second and third commands.

Any comment will be appreciated. Mark as Solution if your query is resolved
__________________
Thanks in Advance
Zahid Haseeb

zahidhaseeb.wordpress.com

mikebounds's picture

You can create resources manually in VCS GUI (or VOM or hagrp & hares command line) and this is what I do and this would replace the cfsmntadm commands.

Mike

UK Symantec Consultant in VCS, GCO, SF, VVR, VxAT on Solaris, AIX, HP-ux, Linux & Windows

If this post has answered your question then please click on "Mark as solution" link below

Zahid.Haseeb's picture

Thanks Mike.

My example Service Group is below....

Application Resource      (Online only on single Nodes)

|

|

cfsmount Resource        (Online on both Nodes)

|

|

cfsvoldg Resource    (Online on both Nodes)

=========================================

- I want to switchover my Application from one node to another but my Switchover button is disable.

- I manually kill my Application it is also not failing over to partner node.

Any comment will be appreciated. Mark as Solution if your query is resolved
__________________
Thanks in Advance
Zahid Haseeb

zahidhaseeb.wordpress.com

mikebounds's picture

The service group containing CVMVolDg and CFSMount is (should be) a parallel service group (the cfsdgadm command will create a parallel group), so parallel groups online on both systems at the same time.  If you application is not parallel and should failover, then you should put your application in a separate failover group dependent on the parallel group.

Mike

UK Symantec Consultant in VCS, GCO, SF, VVR, VxAT on Solaris, AIX, HP-ux, Linux & Windows

If this post has answered your question then please click on "Mark as solution" link below

Zahid.Haseeb's picture

Emm nice.worth it

Any comment will be appreciated. Mark as Solution if your query is resolved
__________________
Thanks in Advance
Zahid Haseeb

zahidhaseeb.wordpress.com

Zahid.Haseeb's picture

Group dependency rule ?

I created online-local firm.

(able to switchover successfully BUT when killing Application/killing the process which is being monitored in cluster, its not failing over and when rebooting the Active system , its failing over)

Any comment will be appreciated. Mark as Solution if your query is resolved
__________________
Thanks in Advance
Zahid Haseeb

zahidhaseeb.wordpress.com

mikebounds's picture

 online-local firm is correct.  You should look at log to why it doesn't failover and why active system is failing over.  

Mike

UK Symantec Consultant in VCS, GCO, SF, VVR, VxAT on Solaris, AIX, HP-ux, Linux & Windows

If this post has answered your question then please click on "Mark as solution" link below

Zahid.Haseeb's picture

emmm I said

killing the application process ID(From Active Node) is not initiating failover BUT rebooting the node(From Active Node) is initiating failing over

The process which I am killing is also mentioned in the process monitor attribute.

Let me see the logs

Any comment will be appreciated. Mark as Solution if your query is resolved
__________________
Thanks in Advance
Zahid Haseeb

zahidhaseeb.wordpress.com

g_lee's picture

Zahid,

Agree with Mike you need to check the logs - may also be worth checking the Application resource that you're testing has Critical=1.

regards,
Grace

If this post has helped you, please vote or mark as solution

SOLUTION
Zahid.Haseeb's picture

Thanks Grace. Thats my mistake. Actually just for checking I did not mark the the Application resource as critical. This is the thing which I forgot to notice

(Marking you post as solution +Thumb) .

Thanks guys

Any comment will be appreciated. Mark as Solution if your query is resolved
__________________
Thanks in Advance
Zahid Haseeb

zahidhaseeb.wordpress.com