HACMP: options for NFS filesystem

[ogniemi]

hi,

there is a cascading cluster built of two nodes on HACMP 4.5. In normal status the primary node has all shared filesystems mounted locally and the second mounts them as a NFS filesystems. The NFS mounts are performed by HACMP as this is configured in the primary resource group in: "Filesystems to be NFS mounted" (there are two resource groups configured in the cluster - one brings primary services up and the other application server services - each started on its "own" node).

The problem is that one of the shared filesystem which is being exported by HACMP on node1 and mounted by HACMP on node2 needs to be mounted with a specific option: noac

Is there a way (not including mounting by script) to tell to HACMP to mount a specific NFS filesystem with a given option? - not standard?

I am not able to change options of the NFS /filesystem locally on each cluster node as there are already stanzas of shared filesystem (/filesystem) to be mounted locally on node2 in case of fall-over of node1.


regards,m.

 

[DSMARWAY]

hi

what do you mean by

'I am not able to change options of the NFS /filesystem locally on each cluster node as there are already stanzas of shared filesystem (/filesystem) to be mounted locally on node2 in case of fall-over of node1.'

are you saying on node2 there is already a mount point with the same name as the one on node 1 ?

one other thing when you created your nfs mount point are there no optionsmake a specifi mount point different to the standard , there should be ?

 

[ogniemi]

As the VG is shared it has been imported on each cluster node. The VG is active/varied-on on one node only at the time. All filesystems a NOT auto-mounted - they are mounted by HACMP (the shared VG is included to resource group configuration and ALL filesystem created in the VG are mounted on the node which requires primary resource group). The other node of the cluster needs exactly the same filesystems (the same mount points) but NFS mounted from primary node. Of course when primary node is going down, NFS filesystem on second node are being unmounted and local filesystems from shared VG are mounted instead.

(the different mount point is not at stake.)

As I mentioned there are no customized options for NFS filesystems to be given in resource group.