[DRBD-user] Restarting DRBD after no-disk-flushes?

[Lars Ellenberg]

On Tue, Apr 14, 2009 at 04:03:05PM -0700, Mike Sweetser - Adhost wrote:
> Hello:
> 
> Is it true that you have to bring the resource down and up after adding
> no-disk-flushes or no-md-flushes to the drbd.conf?  Doing an adjust
> isn't sufficient?

Adjust is supposed to be sufficient to enable those parameters.

Beware of changing disk parameters on a node in Primary role, though,
as that does a detach/attach sequence. And until very recently,
we required a full resync after a detach on a Primary.  Now we lessend
that to a resync of the area covered by the activity log, which still
can be "a lot".

So the general recommendation is do adjust disk parameters while being
in Secondary role, where a detach/attach sequence only leads to a very
quick resync of possibly modified blocks during that timeframe.

Usually, to change disk parameters on both nodes on a live DRBD,
you do either a switchover, or "take it down" for a short period
(in the sense of at least deactivating it into Secondary role).

-- 
: 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.
__
please don't Cc me, but send to list   --   I'm subscribed