[DRBD-user] Many messages drbd_receiver.c:2190: Connectedflags=0x21208

Matthias Weigel matthias.weigel at maweos.de
Fri Mar 23 08:49:06 CET 2007

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


Hello Lars,

what i did before the problem showed up was:
- boot primary
- upgrade primary from 0.7.22 to 0.7.23
- reboot primary
- boot secondary
- upgrade secondary from 0.7.22 to 0.7.23
- reboot secondary
- wait some time
- reboot primary
- now the messages show up on primary, not on secondary
- at this point i sent the mail to this list
- after permanent shutdown of secondary, no more messages on primary.

During all these steps (i simulated upgrading in testenv.), no real 
activity was on the DRBD filesystem. The device is very small (1G). All 
syncs were done very fast and always showed "Consistent" when i checked.

I can somewhat reproduce the problem by just rebooting in the following 
sequence:
- start with secondary+primary shut down
- boot primary
- heartbeat makes primary "Primary".
- boot secondary
- everything is fine now
- "shutdown -r now" on primary
- heartbeat on secondary takes over
- primary comes up, heartbeat makes it "Primary" again
- now when accessing the DRBD-filesystem on primary, the messages start.
Sometimes, a simple "ls" creates the messages. But not every "ls" or 
"find" creates them.
Today, after some 10-15 min, the problem seems to have fixed itself! The 
messages stopped without me doing anything else.

I use ntp on both nodes. Can this be related?
Maybe this is only a problem in VMware?


Thanks for your help

Matthias


Lars Ellenberg schrieb:
> On Thu, Mar 22, 2007 at 02:56:38PM +0100, Matthias Weigel wrote:
>> Hi,
>>
>> when i access a filesystem on top of drbd (with a homemade perl script, 
>> or just with "find ."), i get a lot of these messages on console and syslog:
>>
>> Mar 22 14:39:56 fth1 kernel: drbd0: 
>> /usr/src/drbd-0.7.23/dist/BUILD/drbd-0.7.23/drbd/drbd_receiver.c:2190:Connectedflags=0x21009
>> Mar 22 14:40:29 fth1 kernel: drbd0: 
>> /usr/src/drbd-0.7.23/dist/BUILD/drbd-0.7.23/drbd/drbd_receiver.c:2190:Connectedflags=0x21209
>> Mar 22 14:40:29 fth1 kernel: drbd0: 
>> /usr/src/drbd-0.7.23/dist/BUILD/drbd-0.7.23/drbd/drbd_receiver.c:2190:Connectedflags=0x21208
>> Mar 22 14:40:29 fth1 last message repeated 6 times
>> Mar 22 14:40:34 fth1 kernel: drbd0: 
>> /usr/src/drbd-0.7.23/dist/BUILD/drbd-0.7.23/drbd/drbd_receiver.c:2190:Connectedflags=0x21209
>> Mar 22 14:40:34 fth1 kernel: drbd0: 
>> /usr/src/drbd-0.7.23/dist/BUILD/drbd-0.7.23/drbd/drbd_receiver.c:2190:Connectedflags=0x21208
>> Mar 22 14:40:34 fth1 last message repeated 29 times
>>
>> What does this mean? Is this something serious? The filesystem on top of 
>> drbd still seems to work.
>>
> 
> aparently "__drbd_set_in_sync" is called, even though there is no sync.
> this should not happen, since in "normal operation" no bits are out of
> sync, so nothing could be set in sync.
> 
> you had some sync going on?
> did it terminate ok?
> can you "fix it" by doing
> 	drbdadm disconnect all
> 	# wait for it to go "StandAlone", then
> 	drbdadm connect all
> 
> if you did nothing "stupid", maybe we have some race condition
> or other logic bug for resetting the "SYNC_STARTED" bit somewhere ...
> 
>> cat /proc/drbd
>> version: 0.7.23 (api:79/proto:74)
>> SVN Revision: 2686 build by root at fth1, 2007-03-22 09:06:26
>>  0: cs:Connected st:Primary/Secondary ld:Consistent
>>     ns:1144 nr:8 dw:1152 dr:901 al:1 bm:1 lo:0 pe:0 ua:0 ap:0
>>  1: cs:Connected st:Primary/Secondary ld:Consistent
>>     ns:1 nr:5 dw:6 dr:453 al:0 bm:1 lo:0 pe:0 ua:0 ap:0
>>
>> uname -a
>> Linux fth1 2.6.9-42.0.8.EL #1 Tue Feb 27 14:48:22 CST 2007 i686 i686 
>> i386 GNU/Linux
>>
>> The Linux is latest Whitebox EL4. Primary and Secordary have same 
>> versions in everything and run in VMware.
>>
>>
>> Best Regards
>>
>> Matthias



More information about the drbd-user mailing list