[sheepdog] [PATCH 0/4] bugfix for erasure coding recovery

Hitoshi Mitake mitake.hitoshi at lab.ntt.co.jp
Fri Oct 24 04:10:40 CEST 2014


At Thu, 23 Oct 2014 22:46:19 +0200,
Valerio Pachera wrote:
> 
> 2014-10-23 17:00 GMT+02:00 Hitoshi Mitake <mitake.hitoshi at gmail.com>:
> > Yes, what kind of error message can be found? I think error messages
> > related to failed peer read can still be found. These are known
> > problem which can caused by vdi attr and vdi check, too. But,
> >  Oct 21 12:09:18  ALERT [rw 14158] rollback_vnode_info(117) cannot get epoch 0
> >  Oct 21 12:09:18  ALERT [rw 14158] rollback_vnode_info(118) clients may
> >  see old data
> > this sort of messages would be vanished.
> 
> Yes, they are gone.
> Here's a summary of what I get now.
> 
> Oct 23 22:37:10   INFO [main] recover_object_main(905) object recovery
> progress   1%
> Oct 23 22:37:10   INFO [main] recover_object_main(905) object recovery
> progress   2%
> Oct 23 22:37:10  ERROR [rw 1383] sheep_exec_req(1170) failed No object
> found, remote address: 192.168.10.4:7000, op name: READ_PEER
> Oct 23 22:37:10  ERROR [rw 1383] read_erasure_object(227) can not read
> 7c2b2500000007 idx 0
> Oct 23 22:37:10  ERROR [rw 1382] sheep_exec_req(1170) failed No object
> found, remote address: 192.168.10.4:7000, op name: READ_PEER
> Oct 23 22:37:10  ERROR [rw 1382] read_erasure_object(227) can not read
> 7c2b2500000006 idx 0
> Oct 23 22:37:10  ERROR [rw 1382] sheep_exec_req(1170) failed No object
> found, remote address: 192.168.10.4:7000, op name: READ_PEER
> Oct 23 22:37:10  ERROR [rw 1382] read_erasure_object(227) can not read
> 7c2b2500000006 idx 2
> Oct 23 22:37:10  ERROR [rw 1382] recover_erasure_object(517) failed to
> recover 7c2b2500000006 idx 0
> Oct 23 22:37:10  ERROR [rw 1382] recover_object_work(575) failed to
> recover object 7c2b2500000006
> Oct 23 22:37:10   INFO [main] recover_object_main(905) object recovery
> progress   4%
> <cut>
> 
> grep -c 'rw 1383' /var/lib/sheepdog/sheep.log
> 204
> grep -c 'rw 1382' /var/lib/sheepdog/sheep.log
> 186
> grep -c 'rw 1381' /var/lib/sheepdog/sheep.log
> 168
> grep -c 'rw 1330' /var/lib/sheepdog/sheep.log
> 168
> grep -c 'recover_object_main' /var/lib/sheepdog/sheep.log
> 85
> 
> This with a single vdi of 1G
> (dog vdi create -P test 1G).
> 
> Thank you.

The above error messages will be removed before v1.0, but it remains
in v0.9.0. I think v0.9.1 or later will solve it.

Thanks,
Hitoshi




More information about the sheepdog mailing list