Bug #11481
Updated by Greg Farnum almost 9 years ago
http://pulpito-rdu.front.sepia.ceph.com/teuthology-2015-04-22_11:17:57-fs-hammer-testing-basic-typica/2417/
<pre>2015-04-22 15:36:47.891526 7fde5b0cd700 1 -- 172.20.133.65:6811/16012 <== mon.1 172.20.133.69:6789/0 54 ==== mdsmap(e 23) v1 ==== 867+0+0 (3456897465 0 0) 0x5a88480 con 0x39ef6e0
2015-04-22 15:36:47.891561 7fde5b0cd700 5 mds.-1.0 handle_mds_map epoch 23 from mon.1
2015-04-22 15:36:47.891578 7fde5b0cd700 10 mds.-1.0 my compat compat={},rocompat={},incompat={1=base v0.20,2=client writeable ranges,3=default file layouts on dirs,4=dir inode in separate object,5=mds uses versioned encoding,6=dirfrag is stored in omap,7=mds uses inline data,8=no anchor table}
2015-04-22 15:36:47.891584 7fde5b0cd700 10 mds.-1.0 mdsmap compat compat={},rocompat={},incompat={1=base v0.20,2=client writeable ranges,3=default file layouts on dirs,4=dir inode in separate object,5=mds uses versioned encoding,6=dirfrag is stored in omap,8=no anchor table}
2015-04-22 15:36:47.891587 7fde5b0cd700 10 mds.0.5 map says i am 172.20.133.65:6811/16012 mds.0.5 state up:replay
2015-04-22 15:36:47.891591 7fde5b0cd700 1 mds.0.5 handle_mds_map i am now mds.0.5
2015-04-22 15:36:47.891593 7fde5b0cd700 1 mds.0.5 handle_mds_map state change up:standby --> up:replay
2015-04-22 15:36:47.891594 7fde5b0cd700 10 mds.0.5 set_want_state up:standby -> up:replay
2015-04-22 15:36:47.891597 7fde5b0cd700 1 mds.0.5 replay_start
2015-04-22 15:36:47.891599 7fde5b0cd700 7 mds.0.cache set_recovery_set
2015-04-22 15:36:47.891601 7fde5b0cd700 1 mds.0.5 recovery set is
2015-04-22 15:36:47.891604 7fde5b0cd700 2 mds.0.5 boot_start 0: opening inotable
2015-04-22 15:36:47.891608 7fde5b0cd700 10 mds.0.inotable: load
2015-04-22 15:36:47.894085 7fde5b0cd700 -1 mds/MDSTable.cc: In function 'void MDSTable::load(MDSInternalContextBase*)' thread 7fde5b0cd700 time 2015-04-22 15:36:47.891613
mds/MDSTable.cc: 146: FAILED assert(is_undef())
ceph version 0.94.1-6-g8a58d83 (8a58d83b0d039d2c2be353fee9c57c4e6181b662)
1: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0x7f) [0x966faf]
2: (MDSTable::load(MDSInternalContextBase*)+0x4a6) [0x7b7366]
3: (MDS::boot_start(MDS::BootStep, int)+0x381) [0x5bfc91]
4: (MDS::replay_start()+0x127) [0x5c0ea7]
5: (MDS::handle_mds_map(MMDSMap*)+0x2fb7) [0x5c4107]
6: (MDS::handle_core_message(Message*)+0x26b) [0x5c5bbb]
7: (MDS::_dispatch(Message*)+0x2b) [0x5c626b]
8: (MDS::ms_dispatch(Message*)+0x1e4) [0x5c7704]
9: (Messenger::ms_deliver_dispatch(Message*)+0x77) [0xa5a357]
10: (DispatchQueue::entry()+0x44a) [0xa574fa]
11: (DispatchQueue::DispatchThread::entry()+0xd) [0x94facd]
12: (()+0x7e9a) [0x7fde60197e9a]
13: (clone()+0x6d) [0x7fde5eb5e2ed]
NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.</pre>
This probably is because of our recent changes to MDSTable handling.