Can you please open a ticket in at tracker.ceph.com with this backtrace, and some info about what workload and system config led to this? Are you using erasure coding and/or tiering?
Thanks! sage On Thu, 15 May 2014, Sergey Korolev wrote: > Hello, I have some trouble with OSD. It's crashed with error > > > osd/osd_types.h: 2868: FAILED assert(rwstate.empty()) > > ceph version 0.80.1 (a38fe1169b6d2ac98b427334c12d7cf81f809b74) > 1: (SharedPtrRegistry<hobject_t, > ObjectContext>::OnRemoval::operator()(ObjectContext*)+0x2f5) [0x8dfee5] > 2:(std::tr1::__shared_count<(__gnu_cxx::_Lock_policy)2>::~__shared_count()+0x > 49) [0x66e839] > 3: (ReplicatedPG::OpContext::~OpContext()+0xff) [0x8de93f] > 4: (ReplicatedPG::RepGather::put()+0x37) [0x8deda7] > 5: (C_OSD_RepopCommit::~C_OSD_RepopCommit()+0x24) [0x8defc4] > 6:(ReplicatedBackend::sub_op_modify_reply(std::tr1::shared_ptr<OpRequest>)+0x > 379) [0x967de9] > 7: > (ReplicatedBackend::handle_message(std::tr1::shared_ptr<OpRequest>)+0x3be) > [0x9695ee] > 8: (ReplicatedPG::do_request(std::tr1::shared_ptr<OpRequest>, > ThreadPool::TPHandle&)+0x250) [0x847000] > 9: (OSD::dequeue_op(boost::intrusive_ptr<PG>, > std::tr1::shared_ptr<OpRequest>, ThreadPool::TPHandle&)+0x37c) [0x60e96c] > 10: (OSD::OpWQ::_process(boost::intrusive_ptr<PG>, > ThreadPool::TPHandle&)+0x63d) [0x64068d] > 11: (ThreadPool::WorkQueueVal<std::pair<boost::intrusive_ptr<PG>, > std::tr1::shared_ptr<OpRequest> >, boost::intrusive_ptr<PG> > >::_void_process(void*, ThreadPool::TPHandle&)+0xae) [0x6768fe] > 12: (ThreadPool::worker(ThreadPool::WorkThread*)+0x551) [0xab5721] > 13: (ThreadPool::WorkThread::entry()+0x10) [0xab8760] > 14: (()+0x79d1) [0x7f701bb3a9d1] > 15: (clone()+0x6d) [0x7f701a875b6d] > NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to > interpret this. > > > ceph version 0.80.1 (a38fe1169b6d2ac98b427334c12d7cf81f809b74) > >
_______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
