[DRBD-user] secondary node is inconsistent

Lars Ellenberg Lars.Ellenberg at linbit.com
Mon Jun 26 09:01:07 CEST 2006

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


/ 2006-06-24 12:33:38 +0200
\ Andreas Schader:
> I found out, that when I reboot both nodes with "shutdown -r now" at
> the same time the syncing starts after both are up again and soon
> after that secondary goes back to "Consistent" in /proc/drbd. But I
> can't think that is working as intended that I have to reboot the
> still working primary after a power or network failure of secondary.
> There must be some way how I can get secondary to sync without
> rebooting the primary.

is this a dedicated replication link?

as a workaround for whatever the real problem is with your setup,
try to minimize nfs-activity, and do
# use up some memory, get the caches and stuff to shrink
perl -e '$x = "X" x (500*1024*1024)'
# reconnect
drbdadm connect all

some technical background:
both nodes have to exchange the bitmap, i.e. send the bitmap over,
bitor it together, then store it back on disk...  disk access here in
drbd-0.7 is still synchronouse sector access.  this will take some
considerable amount of time if the io-subsystem is busy doing other
things, depending on your io-latency.  you want to have a look at the
io-scheduler of the lower level devices, too.  anticipatory will not
do you any good with this access pattern.  I'd recommend deadline, with
intelligent storage controllers even noop.


-- 
: 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