[DRBD-user] What situations can trigger inconsistent state?

Lars Ellenberg Lars.Ellenberg at linbit.com
Mon Nov 27 15:42:09 CET 2006

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


/ 2006-11-27 14:44:58 +0300
\ Igor Yu. Zhbanov:
> On Mon, 27 Nov 2006, Lars Ellenberg wrote:
> 
> > / 2006-11-25 23:47:59 +0300
> > \ Igor Yu. Zhbanov:
> > > Hello!
> > > 
> > > I am curious if I use protocol C and initially have drbd in primary/secondary
> > > state, what situations can trigger inconsistent state on primary or
> > > on secondary? (E.g., network problems during long write or resynchronization.)
> > > What should happen to make inconsistency?
> > 
> > 1. after initializing the meta data for the first time,
> >    drbd starts in "Inconsistent".
> > 2. during resync, SyncTarget is Inconsistent
> So, the only way to get inconsistent state on already working system
> is to interrupt synchronization after reconnect?

3. lower level io-error -> detach
   tries to mark itself as inconsistent in meta data.

> Also, I have a question about EXT3 filesystem. Do I need to call ext3fsck
> before each mount in the case of failover cluster with HeartBeat? As I
> understand, drbd consistent state doesn't garantee filesystem consistence?
> Or ext3 always checked automatically each mount?

it does replay its journal on each mount.
this should be sufficient if everything in the stack works correctly.
this is why you use journalling file systems after all.

-- 
: 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.



More information about the drbd-user mailing list