Cool enough, got the log now... last_ack < tid test failing. Some transaction not making it through? I originally intended to build this with -g (rpm), but my build must've been looking at an incorrect/older version of that package. [write 1204224~57344] ack = 0) v1 ==== 106+0+0 (462844183 0 0) 0x7fc19c0aece0 con 0x238ed80 osdc/ObjectCacher.cc: In function 'void ObjectCacher::bh_write_ack(int, sobject_t, loff_t, uint64_t, tid_t)', in thread '0x7fc1ced85710' osdc/ObjectCacher.cc: 673: FAILED assert(ob->last_ack_tid < tid) ceph version 0.25.1 (commit:7f4a161e7fd1bb0d0eab3958b427a3c4c585bc05) 1: (ObjectCacher::bh_write_ack(int, sobject_t, long, unsigned long, unsigned long)+0x6b0) [0x4ef5a0] 2: (ObjectCacher::C_WriteAck::finish(int)+0x5a) [0x4fe30a] 3: (Objecter::handle_osd_op_reply(MOSDOpReply*)+0x93c) [0x4d0d6c] 4: (Client::ms_dispatch(Message*)+0x143) [0x4a4e43] 5: (SimpleMessenger::dispatch_entry()+0x882) [0x454b32] 6: (SimpleMessenger::DispatchThread::entry()+0x1c) [0x44d0ac] 7: (()+0x77e1) [0x7fc1d21b77e1] 8: (clone()+0x6d) [0x7fc1d10a351d] ceph version 0.25.1 (commit:7f4a161e7fd1bb0d0eab3958b427a3c4c585bc05) 1: (ObjectCacher::bh_write_ack(int, sobject_t, long, unsigned long, unsigned long)+0x6b0) [0x4ef5a0] 2: (ObjectCacher::C_WriteAck::finish(int)+0x5a) [0x4fe30a] 3: (Objecter::handle_osd_op_reply(MOSDOpReply*)+0x93c) [0x4d0d6c] 4: (Client::ms_dispatch(Message*)+0x143) [0x4a4e43] 5: (SimpleMessenger::dispatch_entry()+0x882) [0x454b32] 6: (SimpleMessenger::DispatchThread::entry()+0x1c) [0x44d0ac] 7: (()+0x77e1) [0x7fc1d21b77e1] 8: (clone()+0x6d) [0x7fc1d10a351d] *** Caught signal (Aborted) ** in thread 0x7fc1ced85710 ceph version 0.25.1 (commit:7f4a161e7fd1bb0d0eab3958b427a3c4c585bc05) 1: cfuse() [0x553dcc] 2: (()+0xf4c0) [0x7fc1d21bf4c0] 3: (gsignal()+0x35) [0x7fc1d0ff49b5] 4: (abort()+0x175) [0x7fc1d0ff6195] 5: (__gnu_cxx::__verbose_terminate_handler()+0x12d) [0x7fc1d1899aad] 6: (()+0xbcc36) [0x7fc1d1897c36] 7: (()+0xbcc63) [0x7fc1d1897c63] 8: (()+0xbcd5e) [0x7fc1d1897d5e] 9: (ceph::__ceph_assert_fail(char const*, char const*, int, char const*)+0x373) [0x5400c3] 10: (ObjectCacher::bh_write_ack(int, sobject_t, long, unsigned long, unsigned long)+0x6b0) [0x4ef5a0] 11: (ObjectCacher::C_WriteAck::finish(int)+0x5a) [0x4fe30a] 12: (Objecter::handle_osd_op_reply(MOSDOpReply*)+0x93c) [0x4d0d6c] 13: (Client::ms_dispatch(Message*)+0x143) [0x4a4e43] 14: (SimpleMessenger::dispatch_entry()+0x882) [0x454b32] 15: (SimpleMessenger::DispatchThread::entry()+0x1c) [0x44d0ac] 16: (()+0x77e1) [0x7fc1d21b77e1] 17: (clone()+0x6d) [0x7fc1d10a351d] On Fri, Mar 18, 2011 at 4:53 PM, Sage Weil <sage@xxxxxxxxxxxx> wrote: > On Fri, 18 Mar 2011, Brian Chrisman wrote: >> I'm doing some testing via cfuse (technically via samba-over-cfuse). >> I caught the core dump and pulled a stack trace. This is the thread >> that seems to be initiating the core dump. >> I can provide the rest if it'll help identify what's going on. > > If this is reproducible, can you pass --log-file /some/path to cfuse to > get the crash dump (assert error, line number, etc.)? Also, you should > probably build with -g to get the debug symbols (I suspect that's why > you're getting the no symbol table errors). Better yet, if you can > reproduce this with cfuse --log-file foo --debug-client 20 --debug-ms 1 > --debug-objectcacher 20 it'll be easy to squash. > > Also which version? > > Thanks! > sage > > >> > >> My cfuse crashing thread: >> >> Thread 1 (Thread 1696): >> #0 0x00007f14ede7d38b in raise () from /lib64/libpthread.so.0 >> No symbol table info available. >> #1 0x0000000000553fe9 in handle_fatal_signal(int) () >> No symbol table info available. >> #2 <signal handler called> >> No symbol table info available. >> #3 0x00007f14eccb29b5 in raise () from /lib64/libc.so.6 >> No symbol table info available. >> #4 0x00007f14eccb4195 in abort () from /lib64/libc.so.6 >> No symbol table info available. >> #5 0x00007f14ed557aad in __gnu_cxx::__verbose_terminate_handler() () >> from /usr/lib64/libstdc++.so.6 >> No symbol table info available. >> #6 0x00007f14ed555c36 in ?? () from /usr/lib64/libstdc++.so.6 >> ---Type <return> to continue, or q <return> to quit--- >> No symbol table info available. >> #7 0x00007f14ed555c63 in std::terminate() () from /usr/lib64/libstdc++.so.6 >> No symbol table info available. >> #8 0x00007f14ed555d5e in __cxa_throw () from /usr/lib64/libstdc++.so.6 >> No symbol table info available. >> #9 0x00000000005400c3 in ceph::__ceph_assert_fail(char const*, char >> const*, int, char const*) () >> No symbol table info available. >> #10 0x00000000004ef5a0 in ObjectCacher::bh_write_ack(int, sobject_t, >> long, unsigned long, unsigned long) () >> No symbol table info available. >> #11 0x00000000004fe30a in ObjectCacher::C_WriteAck::finish(int) () >> No symbol table info available. >> #12 0x00000000004d0d6c in Objecter::handle_osd_op_reply(MOSDOpReply*) () >> No symbol table info available. >> #13 0x00000000004a4e43 in Client::ms_dispatch(Message*) () >> No symbol table info available. >> #14 0x0000000000454b32 in SimpleMessenger::dispatch_entry() () >> No symbol table info available. >> #15 0x000000000044d0ac in SimpleMessenger::DispatchThread::entry() () >> No symbol table info available. >> #16 0x00007f14ede757e1 in start_thread () from /lib64/libpthread.so.0 >> No symbol table info available. >> #17 0x00007f14ecd6151d in clone () from /lib64/libc.so.6 >> No symbol table info available. >> -- >> 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 >> >> > -- > 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 > -- 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