[DRBD-user] drbd without a partition, for use with xen images

Michael michael.auckland at gmail.com
Sat Jul 26 12:26:42 CEST 2008

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


What about creating image inside drbd?

like mkfs.ext3 /dev/drbd1
mount /dev/drbvd1  /mount-drbd1
move  xen image  to drbd ( mv xen.img /mount-drbd1)
use that image for failover.



On Fri, Jul 25, 2008 at 10:37 PM, Lars Ellenberg
<lars.ellenberg at linbit.com> wrote:
> On Fri, Jul 25, 2008 at 11:49:33AM +0200, Heiko wrote:
>> On Fri, Jul 25, 2008 at 11:07 AM, Robert Verspuy <robert at exa-omicron.nl> wrote:
>> > Michael schreef:
>> >>
>> >> Why not to use loop?
>> >> assing xen image to loop1 ( drbd1 (
>> >> loop2 to drbd2 ... etc...
>> >>
>> Hello,
>>
>> do you mean that I create a Xen Image, than do a "losetup /dev/loop1
>> /VM/my.image"
>> and than configure that loop1 as a disk:
>>
>> on server1 {
>>     device     /dev/drbd0;
>>     disk       /dev/loop1
>>     address    192.168.2.1:7788;
>>     meta-disk  internal;
>>   }
>
> do not do this.
> drbd on top of loop _will_ deadlock sooner or later.
>
> you have been warned.
>
>> or can I just do:
>>
>> on server1 {
>>     device     /dev/drbd0;
>>     disk       /VM/my.image
>>     address    192.168.2.1:7788;
>>     meta-disk  internal;
>>   }
>>
>>
>> >
>> > But then (I think), you can't do a live migration ??
>> > because there's no software making sure that writes to the image from both
>> > server do not write to the same block at the same time. (GFS does that)
>> >
>> > If you don't need live migration, you can use the loop, but when you want to
>> > migrate (in a controlled manner, not failover), you have to do the following
>> > steps:
>> > - On Server A, stop Xen DomU
>> > - On Server A, umount loop1 (drbd1)
>> > - On Server A, set drbd1 from primary to secundairy
>> > - On Server B, set drbd1 from secundairy to primary
>> > - On Server B, mount loop1 (drbd1)
>> > - On Server B, Start Xen VM.
>> >
>> we dont use live migration, for now.
>> We just need a failover of the 1. VM goes offline we need the other
>> one, with actuall data
>>
>> I cant use a real partition because we have some VM on that machine
>> that dont need drbd replication,
>> also some are alone and others just do a mysql replication(inside the VM).
>>
>> cheers
>>
>> .rupertt
>>
>> > Regards,
>> > Robert
>> >
>> > --
>> > *Exa-Omicron*
>> > Patroonsweg 10
>> > 3892 DB Zeewolde
>> > Tel.: 088-OMICRON (66 427 66)
>> > Fax: 088-66 427 69
>> > http://www.exa-omicron.nl
>> > _______________________________________________
>> > drbd-user mailing list
>> > drbd-user at lists.linbit.com
>> > http://lists.linbit.com/mailman/listinfo/drbd-user
>> >
>> _______________________________________________
>> drbd-user mailing list
>> drbd-user at lists.linbit.com
>> http://lists.linbit.com/mailman/listinfo/drbd-user
>>
>
> --
> : Lars Ellenberg                           http://www.linbit.com :
> : DRBD/HA support and consulting             sales at linbit.com :
> : LINBIT Information Technologies GmbH      Tel +43-1-8178292-0  :
> : Vivenotgasse 48, A-1120 Vienna/Europe     Fax +43-1-8178292-82 :
> __
> please don't Cc me, but send to list -- I'm subscribed
> _______________________________________________
> drbd-user mailing list
> drbd-user at lists.linbit.com
> http://lists.linbit.com/mailman/listinfo/drbd-user
>



-- 
--
Michael



More information about the drbd-user mailing list