[linux-lvm] re[2]: Newbie VFS-lock question

[Greg Freemyer]

Dale,

Thanks for the response, Questions below:

 >>  I am still having lvcreate lockups, even though I am no longer calling
 >>  xfs_freeze.  Even stranger, calling xfs_freeze -u causes lvcreate to
 >>  continue, even though I had not called xfs_freeze -f.

 >>  It was my understanding that the VFS-lock patch (or lack thereof) would
 >>  allow the mount step to be reliable, not that it would have any impact
 >>  on lvcreate being able to run to completion.
 >>  [snip]
 >>  ---end original message

 >>  With the VFS lock patch, the calls made inside the lvm layer to freeze and
 >>  unfreeze are functionally identical to the ioctls made by xfs_freeze -f
 >>  and
 >>  xfs_freeze -u.  So you are susceptible to the same sort of lockups. 
 >>  That's
 >>  also why xfs_freeze -u jogged the lvcreate loose.

 >>  If you look through the XFS list archives, you'll find some patches I
 >>  posted
 >>  to help alleviate some of the lockups I had seen, but I've still seen a
 >>  few--generally with multiple snapshots of the same source volume with
 >>  heavy
 >>  write I/O directed to the source volume.

Have your patches been incorporated into the XFS code?

Are you talking about multiple simultaneous snapshots, or creating and destroying them in rapid succession.

The later is what my tests does, but my target environment is just to to a single nightly backup, so my test may be overly stressful.

 >>  One way that should not experience lockups is to use neither xfs_freeze
 >>  nor
 >>  the VFS lock patch, but use writable snapshots.  The snapshot won't be a
 >>  consistent filesystem, but mount it with
 >>  the nouuid option and let it do recovery.  This way may not give you what
 >>  you wanted, but at  least it won't lock up.

 >>  Dale J. Stephenson
 >>  steph at snapserver.com

I'm using a SuSE 2.4.19pre1aa1 based kernel currently, and it is not clear to me that it does or does not have the VFS lock patch.  If I understand you correctly, the fact that I am getting lockups in lvcreate that are resolved by xfs_freeze -u is strong indicator that the patch is present.  Correct?

Would an even better test be to explicitly call xfs_freeze -f to freeze the filesystem, then lvcreate to create a snapshot.  Then check the original filesystem and see if it is now unfroze?  I will try that regardless.

I have never compiled a Linux kernel, but I'm sure I can.

How easy should it be to eliminate the one patch?  Are they typically invoked by some kind of specfile that I can edit and keep the patch from being applied?

I guess I'm off to download the source for my kernel and see if I can figure this stuff out.

Greg Freemyer
Internet Engineer
Deployment and Integration Specialist
Compaq ASE - Tru64 v4, v5
Compaq Master ASE - SAN Architect
The Norcross Group
www.NorcrossGroup.com