Ceph

osd crushed in processing deep scrub
this is the log
ceph version 0.94.6 (e832001feaf8c176593e0325c8298e3f16dfb403)
1: /usr/bin/ceph-osd() [0xa2d315]
2: (()+0xf790) [0x7fbe49e18790]
3: (gsignal()+0x35) [0x7fbe48ae2625]
4: (abort()+0x175) [0x7fbe48ae3e05]
5: (_gnu_cxx::_verbose_terminate_handler()+0x12d) [0x7fbe4939ca5d]
6: (()+0xbcbe6) [0x7fbe4939abe6]
7: (()+0xbcc13) [0x7fbe4939ac13]
8: (()+0xbcd0e) [0x7fbe4939ad0e]
9: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0x57a) [0xadcbba]
10: (FileStore::read(coll_t, ghobject_t const&, unsigned long, unsigned long, ceph::buffer::list&, unsigned int, bool)+0xa8d) [0x90c47d]
11: (ReplicatedBackend::be_deep_scrub(hobject_t const&, unsigned int, ScrubMap::object&, ThreadPool::TPHandle&)+0x2ee) [0x96bf8e]
12: (PGBackend::be_scan_list(ScrubMap&, std::vector<hobject_t, std::allocator<hobject_t> > const&, bool, unsigned int, ThreadPool::TPHandle&)+0x32d) [0x8b14ad]
13: (PG::build_scrub_map_chunk(ScrubMap&, hobject_t, hobject_t, bool, unsigned int, ThreadPool::TPHandle&)+0x1fb) [0x7c6a9b]
14: (PG::replica_scrub(MOSDRepScrub*, ThreadPool::TPHandle&)+0x44a) [0x7c716a]
15: (OSD::RepScrubWQ::_process(MOSDRepScrub*, ThreadPool::TPHandle&)+0x102) [0x6a86f2]
16: (ThreadPool::worker(ThreadPool::WorkThread*)+0x4e6) [0xa325b6]
17: (ThreadPool::WorkThread::entry()+0x10) [0xa34190]
18: (()+0x7a51) [0x7fbe49e10a51]
19: (clone()+0x6d) [0x7fbe48b9893d]
NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.