AIX

Hi

I have a doubt. Our setup is a 2 node HACMP cluster. To create a HACMP file system we do via #smitty hacmp -> cspoc.

Imagine if I have done a mistake and created the file system locally in one of the node using #smitty crfs. How to bring this file system now under HACMP?

depends on your resource group configuration, normally you specify a whole volume group to be shared between your nodes, so your filesystem is mounted / unmounted automatically, even if you don't use cspoc to create it.

but you have to make sure, that the /etc/filesystems entries are on both nodes, set automount to no



now you have different vg time stamps on both nodes, so you can either use extended verification and synchronisation, with automatically correct errors to yes, (resource group must be offline) or the command:

Code:

/usr/es/sbin/cluster/utilities/clupdatevgts volumegroup

on the node where you did not create the fs

I wonder if a "learning import" on the other node wouldn't suffice. I haven't had such a situation in a long, long time, but back in HACMP 4.4 this was the way to go.

I hope this helps.

bakunin

Hi,

Can i use the below command online without offline the RG.

/usr/es/sbin/cluster/utilities/clupdatevgts volumegroup

one of our FS is not able to increase through c-spoc. So we are manually increasing. How to update this in second node online.

How to solve this problem ( not able increase FS cspoc)

I checked my clusters, clupdatevgts seems to be not present in hacmp since at least 5.4

importvg -L works, but has the disadvantage that the vg has to be offline


on node where the vg is concurrent passive online, you could use

Code:

varyoffvg vg
importvg -L vgname hdisk
varyonvg -c -n -P vgname

I tried this out on the standbynode of a running hacmp 5.4 testcluster, and it worked without the cluster complaining about it in any logfile, not even in clstrmgr.debug.long



@sunnybee: whats the error message of cspoc?