Galndalf, regarding this one and previous you told about memory consumption - there are too much PGs, so memory consumption is so high as you are observing. Dead loop of osd-never-goes-up is probably because of suicide timeout of internal queues. It is may be not good but expected.
OSD behaviour is ultimately depends of all kinds of knobs you can change, e.g. I had found recently funny issue that the collection warm-up (e.g. bringing collections to the RAM cache) actually slows down OSD rejoin (typical post-peering I/O delays), comparing with regular situation when collections reading from disk upon OSD launch. On Tue, Apr 29, 2014 at 9:22 PM, Gregory Farnum <g...@inktank.com> wrote: > You'll need to go look at the individual OSDs to determine why they > aren't on. All the cluster knows is that the OSDs aren't communicating > properly. > -Greg > Software Engineer #42 @ http://inktank.com | http://ceph.com > > > On Tue, Apr 29, 2014 at 3:06 AM, Gandalf Corvotempesta > <gandalf.corvotempe...@gmail.com> wrote: >> After a simple "service ceph restart" on a server, i'm unable to get >> my cluster up again >> http://pastebin.com/raw.php?i=Wsmfik2M >> >> suddenly, some OSDs goes UP and DOWN randomly. >> >> I don't see any network traffic on cluster interface. >> How can I detect what ceph is doing ? From the posted output there is >> no way to detect if ceph is recovering or not. Showing just a bunch of >> increasing/decreasing numbers doens't help. >> >> I can see this: >> >> 2014-04-29 12:03:49.013808 mon.0 [INF] pgmap v1047121: 98432 pgs: 241 >> inactive, 33138 peering, 25 remapped, 60067 down+peering, 3489 >> remapped+peering, 1472 down+remapped+peering; 66261 bytes data, 1647 >> MB used, 5582 GB / 5583 GB avail >> >> so what, is it recovering? Is it sleeping ? Why is not recovering ? >> >> http://pastebin.com/raw.php?i=2EdugwQa >> why all OSDs from host osd12 and osd13 are down ? Both hosts are up and >> running. >> _______________________________________________ >> 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 _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
