On Tue, 13 Dec 2016, Jens Rosenboom wrote: > We have seen an issue similar to http://tracker.ceph.com/issues/13594 > on a cluster running 0.94.7, where the OSD fails to start after we > needed the powercycle the server. Can the issue be reopened as > affecting hammer or shall I open a new issue? > > ceph version 0.94.7 (d56bdf93ced6b80b07397d57e3fa68fe68304432) > 1: (ceph::__ceph_assert_fail(char const*, char const*, int, char > const*)+0x8b) [0xbb1fab] > 2: (PG::peek_map_epoch(ObjectStore*, spg_t, unsigned int*, > ceph::buffer::list*)+0x885) [0x7c19b5] > 3: (OSD::load_pgs()+0x9b7) [0x6b9cd7] > 4: (OSD::init()+0x17c7) [0x6bd777] > 5: (main()+0x2a31) [0x6480e1] > 6: (__libc_start_main()+0xf5) [0x7f9711c9cf45] > 7: /usr/bin/ceph-osd() [0x661147] > > Updating to 0.94.9 has not helped. It's unclear from the above with if it is really 13594 or something else. Can you capture a core file (ulimit -c unlimited ; ceph-osd -i NNN -f), install debug symbols (yum install ceph-debuginfo | apt-get install ceph-dbg), gdb, and post the backtrace? 'p values' from the crashing context. Thanks! sage -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
