For the question of OSD failures causing MDS crashes, there are many places where the MDS asserts that OSD operations succeeded (grep the code for "assert(r == 0)") -- we could probably do a better job of handling these, e.g. log the OSD error and respawn rather than assert'ing. John On Sat, Jul 19, 2014 at 11:13 AM, Florent B <florent at coppint.com> wrote: > Hi, > > Is it a known issue ? Has it been fixed in recent Firefly releases ? > > On 07/09/2014 03:21 PM, Yan, Zheng wrote: >> there is memory leak bug in standby replay code, your issue is likely >> caused by it. >> > > _______________________________________________ > ceph-users mailing list > ceph-users at lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |