Ceph

This seems to happen in Jewel as well as in master (with https://github.com/ceph/ceph/pull/7223 reverted):

2016-01-29 12:42:01.194836 7fad2d9cd700  0 log_channel(cluster) log [INF] : 0.0 scrub starts
2016-01-29 12:42:01.194985 7fad2d9cd700  0 log_channel(cluster) log [INF] : 0.0 scrub ok
2016-01-29 12:42:03.195309 7fad2b7ff700  0 log_channel(cluster) log [INF] : 0.2 scrub starts
2016-01-29 12:42:03.209832 7fad2b7ff700 -1 os/bluestore/BlueStore.cc: In function 'virtual int BlueStore::collection_list(coll_t, ghobject_t, ghobject_t, bool, int, std::vector<ghobject_t>*, ghobject_t*)' thread 7fad2b7ff700 time 2016-01-29 12:42:03.195333
os/bluestore/BlueStore.cc: 2804: FAILED assert(k >= start_key && k < end_key)

 ceph version 10.0.2-1367-gc634c6a (c634c6aa232c4d21d2bf237c6b2bb9078f5197c3)
 1: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0x80) [0x562cc27a9c70]
 2: (BlueStore::collection_list(coll_t, ghobject_t, ghobject_t, bool, int, std::vector<ghobject_t, std::allocator<ghobject_t> >*, ghobject_t*)+0x2693) [0x562cc23cc2c3]
 3: (PGBackend::objects_list_partial(hobject_t const&, int, int, std::vector<hobject_t, std::allocator<hobject_t> >*, hobject_t*)+0x658) [0x562cc22a7198]
 4: (PG::chunky_scrub(ThreadPool::TPHandle&)+0x808) [0x562cc21c07d8]
 5: (PG::scrub(unsigned int, ThreadPool::TPHandle&)+0x227) [0x562cc21c1d87]
 6: (OSD::ShardedOpWQ::_process(unsigned int, ceph::heartbeat_handle_d*)+0x6ef) [0x562cc21055ff]
 7: (ShardedThreadPool::shardedthreadpool_worker(unsigned int)+0x734) [0x562cc279a124]
 8: (ShardedThreadPool::WorkThreadSharded::entry()+0x10) [0x562cc279d230]
 9: (()+0x76aa) [0x7fad4ebc76aa]
 10: (clone()+0x6d) [0x7fad4ce5aeed]