Ceph

10/31/2013

05:10 PM Ceph Bug #6684: osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

is it possible you've done any shenanigans like 'ceph pg force_create_pg ...' at some point in the past?
Yes, inde... Ivan Kudryavtsev

11:06 AM Ceph Bug #6684: osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

Any chance I can start these OSD without losing data on them? Ivan Kudryavtsev

10:14 AM Ceph Bug #6684: osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

I also have
2013-11-01 00:13:25.635829 mon.0 [WRN] mon.2 10.252.0.4:6789/0 clock skew 0.870987s > max 0.05s
2013... Ivan Kudryavtsev

10:14 AM Ceph Bug #6684: osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

Here it is, attached. Ivan Kudryavtsev

10/30/2013

11:44 PM Ceph Bug #6684: osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

It's on dumpling node.
root@ceph-osd-1-2:/# ceph -v
ceph version 0.67.4 (ad85b8bfafea6232d64cb7ba76a8b6e8252fa0c7)
Ivan Kudryavtsev

10:17 AM Ceph Bug #6684: osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

Here it is. Ivan Kudryavtsev

09:06 AM Ceph Bug #6684: osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

During recovery some osds crashing and after crash unable to up again, looks like in topic "Domino crash" regarding o... Ivan Kudryavtsev

07:35 AM Ceph Bug #6684: osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

Also it seems, that my meta information is corrupted, I have a lot of images in RBD and now, I don't see any:
rbd:... Ivan Kudryavtsev

06:25 AM Ceph Bug #6684 (Rejected): osd/PGLog.cc: 402: FAILED assert(log.head >= olog.tail && olog.head >= log.tail)

Hello, Guys, I'm in a big trouble. After upgrading Bobtail to cuttlefish I met in a
great troubles, unable to make ... Ivan Kudryavtsev

06/15/2013

03:10 AM Ceph Bug #5365: Massive OSD flaps

I found networking bug (not full connectivity). Ticket could be closed.
The reason was that new osd host was unable ... Ivan Kudryavtsev