[Pacemaker] Split-brain on DRBD + Corosync/Pacemaker

Soni Maula Harriz soni.harriz at sangkuriang.co.id
Thu Dec 20 03:26:15 UTC 2012


On Thu, Dec 20, 2012 at 12:25 AM, Felipe Gutierrez <
felipe.o.gutierrez at gmail.com> wrote:

> Hi Soni,
>
> I did these configurations on my DRBD and the correct recovery of
> split-brain worked well.
>
> http://www.drbd.org/users-guide-8.3/s-configure-split-brain-behavior.html#s-split-brain-notification
>
> I believe that the article you sent to me is about split-brain on Corosync
> and it is different of split-brain of DRBD. I didn't understand well why
> yet.
>

that article discuss about split brain on DRBD, i never heard about split
brain in corosync. Configuring automatic recovery in solving split brain in
DRBD is good, but there is still chance to lose some data. Depending on
your system, i think it is better to have a linux bonding in network, to
make the network link redundant, it's a better solution for network failure.


>
> But at least, my system worked :)
>
> Thanks!
> Felipe
>
>
> On Wed, Dec 19, 2012 at 2:34 PM, Felipe Gutierrez <
> felipe.o.gutierrez at gmail.com> wrote:
>
>> Hi Soni, thanks for reply,
>>
>> I understood that is not possible if I don't have a connection
>> back-to-back (dedicated).
>> But I am thinking to create a script that do that for me.
>> The commands are describe here:
>> http://www.hastexo.com/resources/hints-and-kinks/solve-drbd-split-brain-4-steps
>> And these commands will do the role of an administrator network.
>>
>> Do you think is possible?
>>
>
i don't really understand about scripting, usually i use that command to
manually resolve the split brain, if you think your data is crusial and you
dont want to lose it by depending on DRBD's automatic recovery policy


>
>> Thanks
>> Felipe
>>
>>
>> On Wed, Dec 19, 2012 at 9:41 AM, Soni Maula Harriz <
>> soni.harriz at sangkuriang.co.id> wrote:
>>
>>> cutting the communication link between the two nodes is not a valid
>>> failover scenario. both side will think that other nodes offline and become
>>> primary. and if you reconnect them, the splitbrain will happen. you can
>>> make the communication link redundant between the two nodes. maybe these
>>> articles can help :
>>>
>>> http://www.hastexo.com/blogs/martin/2012/07/11/failover-testing-some-technical-background
>>> http://www.drbd.org/users-guide-8.3/s-prepare-network.html
>>>
>>> On Wed, Dec 19, 2012 at 6:21 PM, Felipe Gutierrez <
>>> felipe.o.gutierrez at gmail.com> wrote:
>>>
>>>> Hi everyone,
>>>>
>>>> I have a scenario that I disconnect my primary from the network and the
>>>> secondary assume, becaming primary. After this, I connect the younger
>>>> primary, and both nodes became secondary(DRBD), or Slave on Pacemaker. It
>>>> is because DRBD on younger Primary is Standalone and Outdated. It is a
>>>> split-brain scenario.
>>>> How do I make to my Pacemaker take care of that?
>>>> I read these
>>>> http://www.drbd.org/users-guide-8.3/s-split-brain-notification-and-recovery.html,
>>>> http://www.drbd.org/users-guide-8.3/s-configure-split-brain-behavior.html#s-automatic-split-brain-recovery-configurationand I think I cannot configure only my DRBD resource. I believe I need to
>>>> do something on my Pacemaker.
>>>> I think Stonith is not a good idea too, because I can reconnect my
>>>> younger primary machine.
>>>>
>>>> Thanks in advance,
>>>> Felipe
>>>> --
>>>> *--
>>>> -- Felipe Oliveira Gutierrez
>>>> -- Felipe.o.Gutierrez at gmail.com
>>>> -- https://sites.google.com/site/lipe82/Home/diaadia*
>>>>
>>>> _______________________________________________
>>>> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>>>> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>>>>
>>>> Project Home: http://www.clusterlabs.org
>>>> Getting started:
>>>> http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>>>> Bugs: http://bugs.clusterlabs.org
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>>> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>>>
>>> Project Home: http://www.clusterlabs.org
>>> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>>> Bugs: http://bugs.clusterlabs.org
>>>
>>>
>>
>>
>> --
>> *--
>> -- Felipe Oliveira Gutierrez
>> -- Felipe.o.Gutierrez at gmail.com
>> -- https://sites.google.com/site/lipe82/Home/diaadia*
>>
>
>
>
> --
> *--
> -- Felipe Oliveira Gutierrez
> -- Felipe.o.Gutierrez at gmail.com
> -- https://sites.google.com/site/lipe82/Home/diaadia*
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20121220/cd566781/attachment.htm>


More information about the Pacemaker mailing list