[sheepdog] [PATCH 0/2] avoid heavy disk I/O in the main thread

[Hitoshi Mitake]

At Wed, 15 Oct 2014 17:26:50 +0900,
tuji wrote:
> 
> Hi,
> 
> > I am going to teset it.
> > And I will check real elapsed time of  purgeing  for our system.
> > It will help to know it is bottoleneck or not.
> 
> I check elapsed time.
> 
> It was  101 sec for about 1TB(263967 object) delete.
> 
> Oct 15 10:58:10   INFO [main] recover_object_main(624) object c7ab8e0001de02 is recovered (263966/263967)
> Oct 15 10:58:11   INFO [main] recover_object_main(624) object c7ab8e0003fe15 is recovered (263967/263967)
> Oct 15 10:58:11   INFO [main] default_cleanup(177) cleaning pruge_state_dir
> Oct 15 10:59:52   INFO [main] default_cleanup(179) cleaning pruge_state_dir done
> 
> It is not moment time. I think we needs mechanism avoid-blocking during
> pruge  state dir.
> 
> Mitake-san,
> 
> Becase our system running on 0.7.x,
> I can not test with your avoid-blocking branch.
> 
> I have tested v0.7.6 + your patch.
> https://github.com/masa-tu/sheepdog/tree/v0.7.6-avoid-blocking

Tuji-san, I updated the avoid-blocking branch for reducing memory
consumption. Could you backport it to 0.7.6 and test again?

Thanks,
Hitoshi