I found the problem. Storage1 dpkg shows ceph is running on 0.80.1 and other 2 are running on 0.80.2. The upgrade follows the same method. I don't understand why. I have tried apt-get update and apt-get upgrade on storage1. How do I bring it to 0.80.2? On Tue, Jul 15, 2014 at 12:30 AM, Richard Zheng <richard at o-matrix.org> wrote: > We used to have ubuntu 12.04 with ceph 0.80.1. The upgrade just tries to > use ubuntu 14.04. We have three servers which run 1 mon and 10 OSDs > each. The other 2 servers are ok. > > Before upgrade we didn't enable upstart and just manually started mon on > all three nodes, e.g. start ceph-mon id=storage<id>. Then we started > ceph-mon after touching done and upstart then run start ceph-all. We used > this way to start mon on storage2 and storage3 successfully. But when we > started mon on storage1, it failed. > > Any suggestion? > > > > On Mon, Jul 14, 2014 at 11:09 PM, Joao Eduardo Luis <joao.luis at inktank.com > > wrote: > >> (re-cc'ing list without log file) >> >> Did you change anything in the cluster aside from ubuntu's version? >> >> Were any upgrades performed? If so, from which version to which version >> and on which monitors? >> >> -Joao >> >> >> On 07/15/2014 03:18 AM, Richard Zheng wrote: >> >>> Thanks Joao. Not sure if the log contains any potential sensitive >>> information... So I didn't reply to the list. >>> >>> On Mon, Jul 14, 2014 at 3:14 PM, Joao Eduardo Luis >>> <joao.luis at inktank.com <mailto:joao.luis at inktank.com>> wrote: >>> >>> On 07/14/2014 11:51 PM, Richard Zheng wrote: >>> >>> Hi, >>> >>> We used to have 0.80.1 on Ubuntu 12.04. We recently upgraded to >>> 14.04. >>> However the mon process doesn't start while OSDs are ok. The >>> ceph-mon >>> log shows, >>> >>> 2014-07-14 12:04:17.034407 7fa86ddcb700 -1 >>> mon.storage1 at 0(electing).__elector(147) Shutting down because I >>> >>> do not >>> support required monitor features: { >>> compat={},rocompat={},__incompat={} } >>> >>> >>> Any suggestions? >>> >>> >>> Set 'debug mon = 10' and 'debug ms = 1' on the monitor, rerun the >>> monitor, have this happen again and drop the log somewhere we can >>> take a look at :) >>> >>> -Joao >>> >>> >>> -- >>> Joao Eduardo Luis >>> Software Engineer | http://inktank.com | http://ceph.com >>> >>> >>> >> >> -- >> Joao Eduardo Luis >> Software Engineer | http://inktank.com | http://ceph.com >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.ceph.com/pipermail/ceph-users-ceph.com/attachments/20140715/5219f6f2/attachment.htm>
 |