Sorry, I am not sure whether it is look ok in your production environment. Maybe you could use the command: ceph tell osd.0 injectargs "-osd_scrub_sleep 0.5" . This command would affect only one osd. If it works fine for some days, you could set for all osd. This is just a suggestion. 2015-04-13 14:34 GMT+08:00 Lindsay Mathieson <lindsay.mathieson@xxxxxxxxx>: > > On 13 April 2015 at 16:00, Christian Balzer <chibi@xxxxxxx> wrote: >> >> However the vast majority of people with production clusters will be >> running something "stable", mostly Firefly at this moment. >> >> > Sorry, 0.87 is giant. >> > >> > BTW, you could also set osd_scrub_sleep to your cluster. ceph would >> > sleep some time as you defined when it has scrub some objects. >> > But I am not sure whether is could works good to you. >> > >> Yeah, that bit is backported to Firefly and can definitely help, however >> the suggested initial value is too small for most people who have scrub >> issues, starting with 0.5 seconds and see how it goes seems to work >> better. > > > > Thanks xinze, Christian. > > Yah, I'm on 0.87 in production - I can wait for the next release :) > > In the meantime, from the prior msgs I've set this: > > [osd] > osd_scrub_chunk_min = 1 > osd_scrub_chunk_max = 5 > osd_scrub_sleep = 0.5 > > > Do the values look ok? is the [osd] section the right spot? > > Thanks - Lindsay > > > > -- > Lindsay -- Regards, xinze _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |