On Tue, May 28, 2013 at 2:35 PM, Giuseppe 'Gippa' Paterno' <[email protected]> wrote: > Hi Greg, >> Do I correctly assume that you don't have any CephFS data in the cluster yet? > The funny thing this was a fresh installation. > Just for your information, ceph-deploy didn't worked for me and I had to > do all the operations manually. > I recreated one of the two ceph clusters with bobtail, same config, and > worked like a charm immediately. > >> If so, I'd just delete your current filesystem and metadata pool, then >> recreate them. >> It should all be in the docs. :) > Although empty, I did the ceph-osd --mkfs again, but with no luck. > Also did the deletion of all the pools and recreated again (with a > simple ceph osd create ....) > Regards, > Giuseppe
Oh, not the OSD stuff, just the CephFS stuff that goes on top. Look at http://www.mail-archive.com/[email protected]/msg00029.html Although if you were re-creating pools and things, I think that would explain the crash you're seeing. -Greg Software Engineer #42 @ http://inktank.com | http://ceph.com _______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
