[DRBD-user] drbd not coming up as primary on failover

Lars Ellenberg lars.ellenberg at linbit.com
Tue Dec 14 01:26:58 CET 2004

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

> > I have drbd 0.7.5

On Mon, Dec 13, 2004 at 12:59:06PM -0500, Musard, Kris wrote:
> I had the same problem. Check that your heartbeat timeouts in
> /etc/ha.d/ha.cf are greater than your drbd timeouts in /etc/drbd.conf.  It
> was caused by my setting the heartbeat timeouts very low to detect failures
> quickly without updating the drbd timeouts.  When my failover occurred drbd
> had not yet given up on the other node and refused to become primary.  You
> can check if that is your problem by waiting for another minute after the
> failover then running drbdsetup from the command line.  If it works  then
> you may have to do some tuning on your drbd timeouts.

from ChangeLog:
0.7.6 (api:77/proto:74)
 * Improvements to the drbddisk script, to do the right thing
   in case Heartbeat is configured with a small timeout than DRBD.

which fixes exactly this point, and uses proper exit codes for the
script as well.

More information about the drbd-user mailing list