Yes. It's awkward and the whole "two weights" thing needs a bit of UI reworking, but it's expected behavior. -Greg Software Engineer #42 @ http://inktank.com | http://ceph.com
On Thu, Apr 10, 2014 at 3:59 PM, Craig Lewis <cle...@centraldesktop.com> wrote: > I've got some OSDs that are nearfull. Hardware is ordered, and I've been > using ceph osd reweight (not ceph osd crush reweight) to keep the cluster > healthy until the new hardware arrives. > > Is it expected behavior that marking an osd down removes the ceph osd > reweight? > > > root@ceph1:~# ceph osd dump | grep ^osd.12 > osd.12 up in weight 0.929291 ... > root@ceph1:~# stop ceph-osd id=12 > ceph-osd stop/waiting > root@ceph1:~# ceph osd out 12 > marked out osd.12. > root@ceph1:~# ceph osd dump | grep ^osd.12 > osd.12 down out weight 0 > root@ceph1:~# start ceph-osd id=12 > ceph-osd (ceph/12) start/running, process 8152 > root@ceph1:~# ceph osd in 12 > marked in osd.12. > root@ceph1:~# ceph osd dump | grep ^osd.12 > osd.12 up in weight 1 > > This bit me when I rebooted a node yesterday. The primary boot disk was > bad, and the boot hung. By the time I got it up, all of that node's OSDs > were down and out. Once they were up and in, there was a lot more > backfilling than I was expecting. > > > > -- > > Craig Lewis > Senior Systems Engineer > Office +1.714.602.1309 > Email cle...@centraldesktop.com > > Central Desktop. Work together in ways you never thought possible. > Connect with us Website | Twitter | Facebook | LinkedIn | Blog > > > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
