[DRBD-user] Automatic split-brain recovery strategies configuration

[Lars Ellenberg]

On Mon, Aug 13, 2007 at 03:36:11AM -0500, Abraham olivares Varela wrote:
> Hi,
> 
> Thanks for your answer , i only hava one last question.

there is no such thing as a "last question" ;-)

> I got two nodes one primary and one secondary like these:
> 
> Node 1 connects to Node 2 by a switched network and Node 1 connects to Node 2 by a crossover cable.
> Consider the following software configuration:
> 
> DRBD communicates over the switched network and Heartbeat (HA) communicates over both networks.
> Now, following these steps I can repeat the case when  i have both nodes in a standalone state which
> require user intervention to reconnect the current PRIMARY DRBD node.
> 
> 1.) Node 1 is DRBD Primary and Node 2 is DRBD Secondary
> 2.) I pull the switched network cable from node 1
> 3.) Node 1 detects the failure and HA moves the resources to
>  node 2

why? because of ping nodes?

> 4.) Node 2 is now StandAlone->Primary/Unknown and node 1 is StandAlone->Secondary/Unknown
> 5.) I reconnect the switched network cable from node 1 and it is at this moment when i got "split-brain".
> 
> my question is, there is any posibility to avoid the "split brain" situation
> when y reconnect the cable from node1?

you can only "avoid" it in 3).  you do so by configuring multiple
heartbeat communication channels, and by using some outdate-peer
handler.  then node 2 would have known that it has outdated-data, and
refused to become primary.

in this scenario, once you actually do 3),
you can no longer "avoid" "split brain situation".
you can only "recover" from it later.

-- 
: Lars Ellenberg                            Tel +43-1-8178292-0  :
: LINBIT Information Technologies GmbH      Fax +43-1-8178292-82 :
: Vivenotgasse 48, A-1120 Vienna/Europe    http://www.linbit.com :
__
please use the "List-Reply" function of your email client.