Project

General

Profile

Actions

Bug #9181

closed

Osd: segv in OpTracker::unregister_inflight_op

Added by Sage Weil over 9 years ago. Updated over 9 years ago.

Status:
Resolved
Priority:
High
Assignee:
Category:
OSD
Target version:
-
% Done:

0%

Source:
Q/A
Tags:
Backport:
Regression:
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

(gdb) bt
#0  0x00007f3dba14ef07 in _dl_map_object_deps (map=map@entry=0x7f3dba3504e8, preloads=preloads@entry=0x0, npreloads=npreloads@entry=0, trace_mode=trace_mode@entry=0, open_mode=open_mode@entry=-2147483648) at dl-deps.c:528
#1  0x00007f3dba155aab in dl_open_worker (a=a@entry=0x7f3da27311e8) at dl-open.c:272
#2  0x00007f3dba150ff4 in _dl_catch_error (objname=objname@entry=0x7f3da27311d8, errstring=errstring@entry=0x7f3da27311e0, mallocedp=mallocedp@entry=0x7f3da27311d0, operate=operate@entry=0x7f3dba1559a0 <dl_open_worker>, args=args@entry=0x7f3da27311e8) at dl-error.c:187
#3  0x00007f3dba1553bb in _dl_open (file=0x7f3db80674de "libgcc_s.so.1", mode=-2147483647, caller_dlopen=<optimized out>, nsid=-2, argc=4, argv=0x7fffd5247928, env=0x2502000) at dl-open.c:661
#4  0x00007f3db8021002 in do_dlopen (ptr=ptr@entry=0x7f3da2731400) at dl-libc.c:87
#5  0x00007f3dba150ff4 in _dl_catch_error (objname=0x7f3da27313e0, errstring=0x7f3da27313f0, mallocedp=0x7f3da27313d0, operate=0x7f3db8020fc0 <do_dlopen>, args=0x7f3da2731400) at dl-error.c:187
#6  0x00007f3db80210c2 in dlerror_run (args=0x7f3da2731400, operate=0x7f3db8020fc0 <do_dlopen>) at dl-libc.c:46
#7  __GI___libc_dlopen_mode (name=name@entry=0x7f3db80674de "libgcc_s.so.1", mode=mode@entry=-2147483647) at dl-libc.c:163
#8  0x00007f3db7ff5c65 in init () at ../sysdeps/x86_64/backtrace.c:52
#9  0x00007f3db9876a90 in pthread_once () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_once.S:103
#10 0x00007f3db7ff5d7c in __GI___backtrace (array=<optimized out>, size=100) at ../sysdeps/x86_64/backtrace.c:103
#11 0x000000000097aaaf in BackTrace (s=0, this=0x7f3da2731730) at ./common/BackTrace.h:19
#12 handle_fatal_signal (signum=11) at global/signal_handler.cc:90
#13 <signal handler called>
#14 0x00000000006a6ee0 in pair (this=0xee1910 <vtable for std::tr1::_Sp_counted_base_impl<DeletingState*, SharedPtrRegistry<spg_t, DeletingState>::OnRemoval, (__gnu_cxx::_Lock_policy)2>+48>) at /usr/include/c++/4.8/bits/stl_pair.h:96
#15 construct (this=<optimized out>, __val=..., __p=0xee1910 <vtable for std::tr1::_Sp_counted_base_impl<DeletingState*, SharedPtrRegistry<spg_t, DeletingState>::OnRemoval, (__gnu_cxx::_Lock_policy)2>+48>) at /usr/include/c++/4.8/ext/new_allocator.h:130
#16 _M_create_node (this=<optimized out>, __x=...) at /usr/include/c++/4.8/bits/stl_tree.h:382
#17 _M_insert_ (__v=..., __p=0x2ff7940, __x=0x0, this=0x29fddc8) at /usr/include/c++/4.8/bits/stl_tree.h:1023
#18 std::_Rb_tree<std::pair<double, std::tr1::shared_ptr<TrackedOp> >, std::pair<double, std::tr1::shared_ptr<TrackedOp> >, std::_Identity<std::pair<double, std::tr1::shared_ptr<TrackedOp> > >, std::less<std::pair<double, std::tr1::shared_ptr<TrackedOp> > >, std::allocator<std::pair<double, std::tr1::shared_ptr<TrackedOp> > > >::_M_insert_unique (this=this@entry=0x29fddc8, __v=...) at /usr/include/c++/4.8/bits/stl_tree.h:1382
#19 0x00000000006a2b0a in insert (__x=..., this=0x29fddc8) at /usr/include/c++/4.8/bits/stl_set.h:463
#20 OpHistory::insert (this=this@entry=0x29fdd98, now=..., now@entry=..., op=...) at common/TrackedOp.cc:42
#21 0x00000000006a3184 in OpTracker::unregister_inflight_op (this=0x29fdd20, i=i@entry=0x4335a40) at common/TrackedOp.cc:131
#22 0x00000000006a3501 in OpTracker::RemoveOnDelete::operator() (this=0x5704038, op=0x4335a40) at common/TrackedOp.cc:250
#23 0x00000000006615a9 in std::tr1::_Sp_counted_base<(__gnu_cxx::_Lock_policy)2>::_M_release (this=0x5704020) at /usr/include/c++/4.8/tr1/shared_ptr.h:141
#24 0x0000000000617a87 in ~__shared_count (this=<synthetic pointer>, __in_chrg=<optimized out>) at /usr/include/c++/4.8/tr1/shared_ptr.h:341
#25 ~__shared_ptr (this=<synthetic pointer>, __in_chrg=<optimized out>) at /usr/include/c++/4.8/tr1/shared_ptr.h:541
#26 ~shared_ptr (this=<synthetic pointer>, __in_chrg=<optimized out>) at /usr/include/c++/4.8/tr1/shared_ptr.h:985
#27 OSD::OpWQ::_process (this=0x29fde28, pg=..., handle=...) at osd/OSD.cc:7722
#28 0x00000000006598dc in ThreadPool::WorkQueueVal<std::pair<boost::intrusive_ptr<PG>, std::tr1::shared_ptr<OpRequest> >, boost::intrusive_ptr<PG> >::_void_process (this=0x29fde28, handle=...) at ./common/WorkQueue.h:190
#29 0x0000000000a4cf01 in ThreadPool::worker (this=0x29fd470, wt=0x379be20) at common/WorkQueue.cc:125
#30 0x0000000000a4ddf0 in ThreadPool::WorkThread::entry (this=<optimized out>) at common/WorkQueue.h:317
#31 0x00007f3db9871182 in start_thread (arg=0x7f3da2733700) at pthread_create.c:312
#32 0x00007f3db7fe538d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111

ubuntu@teuthology:/a/teuthology-2014-08-19_02:30:02-rados-firefly-distro-basic-multi/435292
Actions #1

Updated by Sage Weil over 9 years ago

  • Status changed from New to Need More Info
  • Priority changed from Urgent to High

no log, core isn't giving me good info :(

Actions #2

Updated by Sage Weil over 9 years ago

  • Assignee set to Somnath Roy
Actions #3

Updated by Samuel Just over 9 years ago

  • Status changed from Need More Info to Resolved

I think this got fixed with the other optracker fix?

Actions #4

Updated by Somnath Roy over 9 years ago

Sam,
This core is different and happening on Firefly. The other optracker port should also be backported to Firefly and need to see if it is still happening.

Actions #5

Updated by Somnath Roy over 9 years ago

Somnath Roy wrote:

Sam,
This core is different and happening on Firefly. The other optracker fixes should also be backported to Firefly and need to see if it is still happening.

Actions

Also available in: Atom PDF