On Sun, Jul 5, 2015 at 5:37 AM, Michael Metz-Martini | SpeedPartner GmbH <metz@xxxxxxxxxxxxxxx> wrote: > Hi, > > after larger moves of serveral placement groups we tried to empty 3 of > our 66 osds by slowly setting weight of them to 0 within the crushmap. > > After move completed we're still experiencing a large amount of files > left on that osd-devices. > > For example pg 5.117: > osdmap e56712 pg 5.117 (5.117) -> up [7,59] acting [7,59] > > But there's still a filled current/5.117_head - Directory on osd 22 > > Neither scrubbing nor deep scrubbing seems to cleanup the osd's (I think > that's correct because scrubbing cleans "within pg's". > > Is there a "scrub osd" - Command to automatically (and savely) get rid > of this left files? Is it otherwise save to delete them manually? If the PG is active+clean and the OSD in question isn't listed in its mapping, you're definitely safe to delete it. I suspect that the OSD is simply throttling its deletes down to a slow enough level that it's taking a while to remove all the files (it throttles them in order to try and avoid interrupting client IO, though there's not much point to that in this case). -Greg > > -- > Kind regards > Michael Metz-Martini > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |