Skip to contentSkip to conversation

`pvmove stuck' in RH bugzilla

DISCUSSION Sergey Senozhatsky 2 participants 3 messages Jul 6, 2011 MD RAID

lore
Share & tools
lore
#0 lore raw

Hello,

I think we just trapped into the same trouble with pvmove as described in

https://bugzilla.redhat.com/show_bug.cgi?id=602516
https://bugzilla.redhat.com/show_bug.cgi?id=706036
etc.

Comment says, that

Quoted 5 lines

I found a series of problems with pvmove which I have fixed in the upstream repository. However, I would like to wait some time to see that no unintended side-effects show up before seeing the change back-ported to RHEL. An audit for similar bugs found a problem in lvremove, which I have also fixed. lvconvert still needs to be audited.

Alasdair, could you please point are those patches in vanilla kernel (3.0-rc*) or linux-next, or some RH internal tree?
This is not critical, though I'd like to try out patches.

Thanks,
Sergey

#1 lore raw
Quoted 7 lines · Sergey Senozhatsky

I think we just trapped into the same trouble with pvmove as described in
https://bugzilla.redhat.com/show_bug.cgi?id=602516
https://bugzilla.redhat.com/show_bug.cgi?id=706036

Alasdair, could you please point are those patches in vanilla kernel (3.0-rc*) or linux-next, or some RH internal tree?
This is not critical, though I'd like to try out patches.

This is a userspace bug not a kernel one.
There is a workaround, which is to perform the pvmove one LV at a time using the --name option. (For each LV, run pvmove -n <lvname>.)

The patched code is in the upstream lvm2 repository:
git://sources.redhat.com/git/lvm2
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/doc/?cvsroot=lvm2
http://sourceware.org/lvm2/

It was only last night that the code incorporating these changes passed our public testsuite for the first time, so I think it is still premature to be using this on production systems. But this means I hope to be in a position to make the first development release containing a fix for the problems quite soon. (I added assertions to catch other instances of the related problems - it was more than just pvmove affected.)

Alasdair

--
lvm-devel mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/lvm-devel

#2 lore raw
Quoted 18 lines · Alasdair G Kergon
Quoted 7 lines · Sergey Senozhatsky

I think we just trapped into the same trouble with pvmove as described in
https://bugzilla.redhat.com/show_bug.cgi?id=602516
https://bugzilla.redhat.com/show_bug.cgi?id=706036

Alasdair, could you please point are those patches in vanilla kernel (3.0-rc*) or linux-next, or some RH internal tree?
This is not critical, though I'd like to try out patches.

This is a userspace bug not a kernel one.
There is a workaround, which is to perform the pvmove one LV at a time using the --name option. (For each LV, run pvmove -n <lvname>.)

The patched code is in the upstream lvm2 repository:
git://sources.redhat.com/git/lvm2
http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/doc/?cvsroot=lvm2
http://sourceware.org/lvm2/

Quoted 8 lines

It was only last night that the code incorporating these changes passed our public testsuite for the first time, so I think it is still premature to be using this on production systems. But this means I hope to be in a position to make the first development release containing a fix for the problems quite soon. (I added assertions to catch other instances of the related problems - it was more than just pvmove affected.)

Thanks a lot for your reply.
We don't have a production machines here (just internal servers), so I guess I could give it a try.

Best,
Sergey