[DRBD-user] [drbd] Rsync secondary after failure

Corey Edwards tensai at zmonkey.org
Thu May 12 02:56:30 CEST 2005

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


On Wed, 2005-05-11 at 18:40, Xavier Grangier wrote:
> Hello all,
> 
> I'm new to the list :) I've benn using drbd on my company lan since 
> february, things vent well for a while.
> Lately I had an hard disk failure on my secondary, the disk is dead an I 
> had to change it for a new one.
> 
> I have trouble to resync the primary and the secondary. I'm not sure 
> about the right procedure, here is wat I did:
> 
> - change the hardrive
> - then mkfs.ext3 on  /dev/hda4 on the secondary (I guess the mistake is 
> somewhere here)
> - then restart drbd on primary and secondary.

The fatal error was in step 2. You should never make changes directly to
the underlying device. Why exactly it failed, I can't tell you as I'm a
relative newbie myself. But the proper sequence would have been to
simply leave out step two. DRBD will detect that the secondary is
invalid and initiate a full sync from the master.

Now that you've got it in this state, I think you'll want to invalidate
the secondary. On bip2, run the command "drbdadm invalidate drbd0". That
will force a FULL sync from backup.

I also don't believe it's necessary to restart DRBD on the primary. It
will automagically detect the new peer as soon as it starts as long as
the configuration has not changed.

Corey

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20050511/e14ff25e/attachment.pgp>


More information about the drbd-user mailing list