[DRBD-user] DRBD V8.4.2 Re-sync after service migration takes long time

Lars Ellenberg lars.ellenberg at linbit.com
Fri Feb 15 10:00:42 CET 2013

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

On Thu, Feb 14, 2013 at 03:51:30PM +0000, Matchett, John wrote:
> My configuration is Pri/Pri  DRBD Cluster with GFS2/CLVM  and
> Postgres/Zabbix    If I kill my Active node,  the  Standby node
> migrates services  quickly and becomes the new Active Node.  However,
> once the old Active node reboots as a new Standby  node,  it takes 5-7
> minutes to re-sync the DRBD partition.    .
> If only  the   Standby node reboots and there is no service migration,
> the Re-sync  only takes a few seconds.  Why does the resync take
> longer following a service migration to standby node  vs when there's
> no service migration with only the standby recovery?
> I did notice the size of the resync 29M  is smaller with just a
> Standby node restart than it is when the Primary is killed, and
> restarts as Standby (4G)

Why would you attach JPEGs of text output,
instead of pasting the text itself?

After a Primary crash,
the area covered by the activity log needs to be resynced.
After a clean shutdown, there is no need for that.

See the DRBD User's Guide, "The Activity Log"

: Lars Ellenberg
: LINBIT | Your Way to High Availability
: DRBD/HA support and consulting http://www.linbit.com

DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.

More information about the drbd-user mailing list