[DRBD-user] DRBD 8.0.2 ASSERT( atomic_read

Lars Ellenberg lars.ellenberg at linbit.com
Tue Apr 24 12:41:34 CEST 2007

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


On Tue, Apr 24, 2007 at 06:06:09PM +0800, Marji Cermak wrote:
> Hello Guys,
> I'm running DRBD 8.0.2 compiled on Centos5.
> 
> I realised my iSCSI target exporting a chunk of a drbd device was timeouting
> because I was too optimistic and had the
> syncer - rate variable too big (an easy mistake to make :-)
> 
> After I set the syncer to { rate 5M; } and restarted drbd on both ends, I could
> see many of these asserts in my log:
> 
> Apr 24 09:14:47 nas02 kernel: drbd22: ASSERT( atomic_read(&mdev->local_cnt) )
> in /usr/src/drbd-8.0.2/drbd/drbd_actlog.c:840
> Apr 24 09:14:47 nas02 kernel: drbd22: ASSERT( atomic_read(&mdev->local_cnt)>=0
> ) in /usr/src/drbd-8.0.2/drbd/drbd_int.h:1768
> Apr 24 09:14:47 nas02 kernel: drbd22: ASSERT( atomic_read(&mdev->local_cnt) )
> in /usr/src/drbd-8.0.2/drbd/drbd_actlog.c:840
> Apr 24 09:14:47 nas02 kernel: drbd22: ASSERT( atomic_read(&mdev->local_cnt)>=0
> ) in /usr/src/drbd-8.0.2/drbd/drbd_int.h:1768
> Apr 24 09:15:18 nas02 last message repeated 3938 times
> Apr 24 09:16:19 nas02 last message repeated 10578 times

not good.
aparently we get the housekeeping wrong sometimes :(

> No other drbd device has this ASSERT problem - just this one which has been
> syncing (and will be for another 10 hours or so):

> [root at nas02 ~]# cat /proc/drbd
> 22: cs:SyncTarget st:Secondary/Secondary ds:Inconsistent/UpToDate C r---
>     ns:0 nr:568101024 dw:568101024 dr:0 al:0 bm:34674 lo:-76 pe:0 ua:0 ap:0
                                                        ^^^^^^^

this lo: counter is "strictly positive". well. it should be, anyways.

the good news is, as long as you don't get any real io-errors from the
lower level devices, the only real impact is that it floods your logs.

for now, only way to fix this is probably to unconfigure/reconfigure it.
drbdadm down drbd22; drbdadm up drbd22

it _may_ get stuck during the "unconfigre", though, waiting for "lo:" to
reach zero... oh well :(

-- 
: Lars Ellenberg                            Tel +43-1-8178292-0  :
: LINBIT Information Technologies GmbH      Fax +43-1-8178292-82 :
: Vivenotgasse 48, 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