Disk Enclosures
1753774 Members
7070 Online
108799 Solutions
New Discussion юеВ

Re: Extending Filesystem on FC60 cluster

 
SOLVED
Go to solution
Nyree Johns
Occasional Contributor

Extending Filesystem on FC60 cluster

Hi,

I've got to extend a filesystem on our FC60 cluster by 3Gb. It's in LUN0 (total capacity 17Gb, filesystems so far 7Gb). The LUN has 2 disks, the primary being /dev/dsk/c9t9d0 and it's raid 1. Can I just run extendfs as normal?

Many thanks.
4 REPLIES 4
Insu Kim
Honored Contributor

Re: Extending Filesystem on FC60 cluster

Unmount the filesystem, extend logical volume as required and finally do extendfs if you don't have OnlineJFS installed.

If you post this subject to "General-> hpux", you will get more responses.

Best regards,
Never say "no" first.
Nyree Johns
Occasional Contributor

Re: Extending Filesystem on FC60 cluster

Hi,

Thanks for the response. I don't know a great deal about the arrays, and so wasn't sure whether a different command set was used with volume/fs extensions or not, which is why I posted it here rather than hp-ux. I take it from your answer though that I can treat it like a normal volume group /filesystem?

Many thanks
Bill McNAMARA_1
Honored Contributor
Solution

Re: Extending Filesystem on FC60 cluster

Yes for sure it's the same behaviour as any other disk device. LVM in any case is logical rather than physical partitioning, so even if there was a difference lvm would hide the physical disk from you..

BUT, having said that, extending a logical volume that is managed by MC/SG is not as straight forward as lvextend. You must make sure that the host that extends the lvol has exclusive access to the vg, and you will have to make sure both hosts know about changes to the vg.

Ask in the serviceguard forum!
I'll try to root out my serviceguard cookbooks in the meantime..

Later,
Bill
It works for me (tm)
Nyree Johns
Occasional Contributor

Re: Extending Filesystem on FC60 cluster

Bill,

Thanks very much. I went off to the serviceguard forum and found the exact same question. As I'm only extending an existing filesystem I don't need to do anything else with the other nodes, so it's a straightforward job. Thanks very much.