[DRBD-user] Re: OOPS with drbd, lvm and XFS on debian xen 2.6.16

Igor Morgado igormorgado.listas at gmail.com
Thu Jul 6 18:59:15 CEST 2006

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


Lars..

My environment "needs" of drbd8. Because live migration works flawlessly
with pri/pri resources. Using sec/pri I need to do a big trip to achieve
this.

sleep the vm
put local disk as secondary
copy the vm state to remote server
put remote disk as primary
load the vm state.

This procedure takes 20-30 seconds to complish. I can even keep the session
(because tcp/ip timeout)

With pri/pri I do the following:
put remote disk as primary
start xen live migration
put local disk as secondary

The procedure takes 1-3 seconds, and I have only 1 second of delay in
session (most of users cannot even notice the migration).


My question is:

Will be any compatibility with drbd7 and 8? there is a way to migrate drbd7
metadata to 8 without data loss?

Best regards




On 7/6/06, Lars Ellenberg <Lars.Ellenberg at linbit.com> wrote:
>
> / 2006-07-05 16:30:10 -0300
> \ Igor Morgado:
> > Sorry to touch in it again. But I was simply ignored, is this report
> erroneous?
> > May I add some information to get some kind of answer about this? There
> is
> > other mail list to report this?
> >
> > Because Im really scared I lost ALL DATA from this volume after this
> happened.
> > It was a test volume but I want to put this over real systems.
> >
> > Any tips.
>
> if you are going production system, don't use developement versions of
> drbd,
> use the stable branch (0.7.x, x == 20 currently).
>
> if you want to help in the developemnt process of drbd 8,
> please try to use the most recent svn checkout
> svn co http://svn.drbd.org/drbd/trunk  drbd-8-svn
>
>
> if you get problems with some "special" kernel, please try to reproduce
> with "vanilla" first, to see if it is related to the special kernel, or
> some more generic problem.
> > On 7/3/06, Igor Morgado <igormorgado.listas at gmail.com> wrote:
> >
> >     This happned after a drbdadm resize in my resource (I had extended
> the
> >     device with lvm), on a debian gnu/linux (the oops is on the end of
> report)
> >
> >     magratea:~# uname -a
> >     Linux magratea 2.6.16-2-xen-686 #1 SMP Tue May 23 05:46:45 CEST 2006
> i686
> >     GNU/Linux
> >
> >
> >     Im running LVM on top of DRBD
> >
> >
> >     magratea:~# lvm version
> >       LVM version:     2.01.04 (2005-02-09)
>
> I think you _must_ use at least 2.02 here with recent kernels.
>
> >       Library version: 1.01.00-ioctl (2005-01-17)
> >       Driver version:  4.5.0
>
> so this should be more like
>   LVM version:     2.02.05 (2006-04-21)
>   Library version: 1.02.05 (2006-04-19)
>   Driver version:  4.5.0
>
> >     drbd version is:
> >     Version: 8.0pre3 (api:82)
> >     SVN Revision: 2169 build by root at magratea, 2006-06-29 17:09:10
>
> try again with recent svn checkout.
>
> double check the kernel stack size.
> 4K is likely to be too small and will overflow.
> this is not neccessary the case here, though. could be a drbd bug in the
> drbd resize code that got already fixed in svn.
>
> --
> : Lars Ellenberg                                  Tel +43-1-8178292-0  :
> : LINBIT Information Technologies GmbH            Fax +43-1-8178292-82 :
> : Schoenbrunner Str. 244, A-1120 Vienna/Europe   http://www.linbit.com :
> __
> please use the "List-Reply" function of your email client.
> _______________________________________________
> drbd-user mailing list
> drbd-user at lists.linbit.com
> http://lists.linbit.com/mailman/listinfo/drbd-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20060706/e04e6555/attachment.htm>


More information about the drbd-user mailing list