> Op 12 juli 2016 om 8:47 schreef Christian Balzer <chibi@xxxxxxx>: > > > > Hello, > > On Tue, 12 Jul 2016 08:39:16 +0200 (CEST) Wido den Hollander wrote: > > > Hi, > > > > I am upgrading a 1800 OSD cluster from Hammer 0.94.5 to 0.94.7 prior to going to Jewel and while doing so I see the monitors being flooded with these messages: > > > Google is your friend (and so is the NSA): > --- > http://www.spinics.net/lists/ceph-devel/msg30450.html > --- > Thanks! I was searching, but never found that thread. Well, not that post in that thread. The messages in my 'ceph -w' are still 20 minutes behind currently. Logging about 08:39 while it's 08:56 here right now. Wido > That's also one of the reasons that despite only having a fraction of your > or Dan's OSDs I'm not upgrading to 0.94.7... > > Christian > > > 2016-07-12 08:28:12.919748 osd.1200 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:12.921943 osd.1338 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:12.923814 osd.353 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:12.939370 osd.1200 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:12.941482 osd.1338 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:12.960100 osd.1338 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:12.979404 osd.1338 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:13.012463 osd.353 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:13.039417 osd.353 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:13.079893 osd.353 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:13.111176 osd.575 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:13.135279 osd.353 [WRN] failed to encode map e130549 with expected crc > > 2016-07-12 08:28:13.144697 osd.575 [WRN] failed to encode map e130549 with expected crc > > > > This just goes on and on. The flood of messages cause the monitors to start consuming a bit of CPU which makes the cluster operate slower. > > > > I am restarting the OSDs slowly and when I stop doing so the messages disappear and the cluster operates just fine. > > > > I know that the messages pop up due to a version mismatch, but is there any way to suppress them? > > > > Wido > > _______________________________________________ > > ceph-users mailing list > > ceph-users@xxxxxxxxxxxxxx > > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > > > > -- > Christian Balzer Network/Systems Engineer > chibi@xxxxxxx Global OnLine Japan/Rakuten Communications > http://www.gol.com/ _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |