[DRBD-user] split brain - why does it not work out?

Raoul Bhatia [IPAX] r.bhatia at ipax.at
Thu Oct 16 14:29:42 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.


Florian Haas wrote:
>> To resolve split brain, I usually try the following:
>>
>> 1- Disconnect both resources on both nodes (i.e. drbdadm disconnect resource)
>> 2- Mark both nodes as secondary
> 
> Not necessary; you need to do this only on one node (the one you
> designate as your split brain victim).
> 
>> 3- on the node you want as your primary do: drbdadm -- --overwrite-data-of-peer primary resource
> 
> No, no, no, NO. You don't use this when resolving split brain.
> 
>> OR
>> On the actual secondary do: drbadm -- --discard-my-data connect resource
> 
> Better.

thank you - i was about to write something similar :)

>> This should "most of the time" prevent split brains....
> 
> This does not prevent split brain. At all. This _recovers_ from split
> brain. What you do to prevent split brain is to use either
> resource-level or node level fencing (i.e. dopd or -- if you don't like
> dopd -- STONITH).

dopd and stonith are both running. stonith would not prevent this issue
as it is the direct result from a hard-reset. i would "expect" dopd to
fix this but somehow it did not. is dopd periodically trying to outdate
the peer, or only once?

cheers,
raoul
-- 
____________________________________________________________________
DI (FH) Raoul Bhatia M.Sc.          email.          r.bhatia at ipax.at
Technischer Leiter

IPAX - Aloy Bhatia Hava OEG         web.          http://www.ipax.at
Barawitzkagasse 10/2/2/11           email.            office at ipax.at
1190 Wien                           tel.               +43 1 3670030
FN 277995t HG Wien                  fax.            +43 1 3670030 15
____________________________________________________________________




More information about the drbd-user mailing list