HACMP filesystems

[SkinDvr]

We have a client with several "Shared File Systems" on a two cluster node. They added another filesystem on the correct VG but didn't create it shared, so nothing mounted after a reboot.

Is there a way to make that filesystem shared across the cluster? Or do we need to create a new one, shutdown and copy over to it.

Thanks!
brian

 

[ggauthier]

If I understand this correctly:
• You have a two node HACMP cluster.
• There is a volume group that is in a resource group that is shared between the systems, but is only varied-on on one system at a time.
• There are filesystems in this volume group
• Your client created a new filesystem in this existing volume group
• The new filesystem does not mount after reboot.

The filesystem should be available to mount manually, and you won't have to re-create or copy anything.

You should get a maintenance window, add that filesystem to the resource group, synchronize the cluster, and test the failover.

You may want to consider revoking privs to your client.

gg

 

[mrn]

The easy way is to use GPFS see

http://www.redbooks.ibm.com/abstracts/sg246035.html

Mike

"A foolproof method for sculpting an elephant: first, get a huge block of marble, then you chip away everything that doesn't look like an elephant.