[prev in list] [next in list] [prev in thread] [next in thread] 

List:       lustre-announce
Subject:    Re: [Lustre-discuss] resize OST backing device
From:       Phil Schwan <phil () clusterfs ! com>
Date:       2005-01-03 17:28:21
Message-ID: BDFEEAE5.6813%phil () clusterfs ! com
[Download RAW message or body]

Hi Jeffrey--

On 12/29/2004 19:18, Jeffrey W. Baker wrote:
> 
> If an OST's backing device is an LVM logical volume, and a physical
> volume is added to that LV, will it be possible to resize the OST to
> take up the new space?

Almost...

You can resize the LV, and if you resize the ext3 volume, Lustre will have
no problem with that.

The only issue is that I don't think the ext3 resizer has been updated to
handle 100% of our improvements to the ext3 disk format.  Andreas Dilger is
the resizer expert, so I'll let him correct me if I'm wrong.

We'll get to this eventually, I'm sure, or perhaps someone will beat us to
it.  I doubt that the general ext3 community will do it until they've
finished evaluating all of our patches.  If this is very important to you, a
small contract can probably be arranged.

> Is it possible to do this while the OST is
> running?  I noticed a few online resize patches for ext3 came across the
> l-k mailing list, but I have no idea if they are in Lustre kernels, nor
> how to invoke a resize (it appears from the patch that one calls mke2fs
> -R resize=$blocks)

The online resizing patches are only in very new 2.6 kernels -- kernels
newer than we currently support.  But subject to the same caveat as above,
when the resizer supports our changes, online resizing will just work as you
would expect.

Hope that helps--

-Phil


_______________________________________________
Lustre-discuss mailing list
Lustre-discuss@lists.clusterfs.com
https://lists.clusterfs.com/mailman/listinfo/lustre-discuss
[prev in list] [next in list] [prev in thread] [next in thread]