Negative degradation?

Negative degradation?

[Roald van Loon]

Came across it this morning when booting my development environment,
has anyone seen this before?

It's with 0.67-rc2;

2013-07-30 08:09:09.230349 mon.0 [INF] pgmap v4172: 1216 pgs: 123
active, 219 active+clean, 161 active+clean+replay, 713 peering; 241 MB
data, 630 MB used, 5483 MB / 6114 MB avail; -2/588 degraded (-0.340%)
2013-07-30 08:09:11.967368 mon.0 [INF] pgmap v4173: 1216 pgs: 187
active, 669 active+clean, 161 active+clean+replay, 199 peering; 241 MB
data, 631 MB used, 5482 MB / 6114 MB avail; -2/588 degraded (-0.340%)
2013-07-30 08:09:12.983522 mon.0 [INF] pgmap v4174: 1216 pgs: 64
active, 991 active+clean, 161 active+clean+replay; 241 MB data, 626 MB
used, 5487 MB / 6114 MB avail
2013-07-30 08:09:16.608991 mon.0 [INF] pgmap v4175: 1216 pgs: 1055
active+clean, 161 active+clean+replay; 241 MB data, 626 MB used, 5487
MB / 6114 MB avail
2013-07-30 08:09:47.985946 mon.0 [INF] pgmap v4176: 1216 pgs: 1117
active+clean, 99 active+clean+replay; 241 MB data, 626 MB used, 5487
MB / 6114 MB avail
2013-07-30 08:09:52.457863 mon.0 [INF] pgmap v4177: 1216 pgs: 1204
active+clean, 12 active+clean+replay; 241 MB data, 626 MB used, 5487
MB / 6114 MB avail
2013-07-30 08:09:53.488936 mon.0 [INF] pgmap v4178: 1216 pgs: 1216
active+clean; 241 MB data, 626 MB used, 5487 MB / 6114 MB avail

Re: Negative degradation?

[David McBride]

On 30/07/13 09:20, Roald van Loon wrote:
> Came across it this morning when booting my development environment,
> has anyone seen this before?
> 
> It's with 0.67-rc2;
> 
> 2013-07-30 08:09:09.230349 mon.0 [INF] pgmap v4172: 1216 pgs: 123
> active, 219 active+clean, 161 active+clean+replay, 713 peering; 241 MB
> data, 630 MB used, 5483 MB / 6114 MB avail; -2/588 degraded (-0.340%)

I saw similar negative values when updating my test OSD host to 0.67-rc2
from 0.66 this morning; however, I've seen such negative values being
reported at various times over a number of past releases.

I suspect there's a defect in signed-ness handling somewhere, or similar
— however, it appears to be cosmetic and disappears when the OSD replay
completes.

(I've likewise seen impossible values for other statistics being
reported — a while back I saw multiple EBs being reported as being
read/written during a resync between 4 disks, for example.  I suspect
this is another manifestation of a similar problem.)

Cheers,
David
-- 
David McBride <dwm37@cam.ac.uk>
Unix Specialist, University Computing Service
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html