On Sun, Oct 23, 2022 at 11:04 PM can zhu <zhucan....@gmail.com> wrote:
>
> crash info:
>
> {
> "backtrace": [
> "/lib64/libpthread.so.0(+0x12ce0) [0x7f82e87cece0]",
> "(BlueStore::Onode::put()+0x1a3) [0x55bd21a422e3]",
> "(std::_Hashtable<ghobject_t, std::pair<ghobject_t const,
> boost::intrusive_ptr<BlueStore::Onode> >,
> mempool::pool_allocator<(mempool::pool_index_t)4, std::pair<ghobject_t
> const, boost::intrusive_ptr<BlueStore::Onode> > >,
> std::__detail::_Select1st, std::equal_to<ghobject_t>,
> std::hash<ghobject_t>, std::__detail::_Mod_range_hashing,
> std::__detail::_Default_ranged_hash, std::__detail::_Prime_rehash_policy,
> std::__detail::_Hashtable_traits<true, false, true> >::_M_erase(unsigned
> long, std::__detail::_Hash_node_base*,
> std::__detail::_Hash_node<std::pair<ghobject_t const,
> boost::intrusive_ptr<BlueStore::Onode> >, true>*)+0x68) [0x55bd21af7bc8]",
> "(BlueStore::OnodeSpace::_remove(ghobject_t const&)+0x29b)
> [0x55bd21a420eb]",
> "(LruOnodeCacheShard::_trim_to(unsigned long)+0xeb)
> [0x55bd21afce6b]",
> "(BlueStore::OnodeSpace::add(ghobject_t const&,
> boost::intrusive_ptr<BlueStore::Onode>&)+0x49d) [0x55bd21a42dfd]",
> "(BlueStore::Collection::get_onode(ghobject_t const&, bool,
> bool)+0x46a) [0x55bd21a86e1a]",
> "(BlueStore::_txc_add_transaction(BlueStore::TransContext*,
> ceph::os::Transaction*)+0x1124) [0x55bd21aab2e4]",
>
> "(BlueStore::queue_transactions(boost::intrusive_ptr<ObjectStore::CollectionImpl>&,
> std::vector<ceph::os::Transaction, std::allocator<ceph::os::Transaction>
> >&, boost::intrusive_ptr<TrackedOp>, ThreadPool::TPHandle*)+0x316)
> [0x55bd21ac76d6]",
> "(non-virtual thunk to
> PrimaryLogPG::queue_transactions(std::vector<ceph::os::Transaction,
> std::allocator<ceph::os::Transaction> >&,
> boost::intrusive_ptr<OpRequest>)+0x58) [0x55bd2170b878]",
>
> "(ReplicatedBackend::do_repop(boost::intrusive_ptr<OpRequest>)+0xeb0)
> [0x55bd218fdff0]",
>
> "(ReplicatedBackend::_handle_message(boost::intrusive_ptr<OpRequest>)+0x267)
> [0x55bd2190e357]",
> "(PGBackend::handle_message(boost::intrusive_ptr<OpRequest>)+0x52)
> [0x55bd2173ed52]",
> "(PrimaryLogPG::do_request(boost::intrusive_ptr<OpRequest>&,
> ThreadPool::TPHandle&)+0x5de) [0x55bd216e268e]",
> "(OSD::dequeue_op(boost::intrusive_ptr<PG>,
> boost::intrusive_ptr<OpRequest>, ThreadPool::TPHandle&)+0x309)
> [0x55bd21569fc9]",
> "(ceph::osd::scheduler::PGOpItem::run(OSD*, OSDShard*,
> boost::intrusive_ptr<PG>&, ThreadPool::TPHandle&)+0x68) [0x55bd217c8e78]",
> "(OSD::ShardedOpWQ::_process(unsigned int,
> ceph::heartbeat_handle_d*)+0xc28) [0x55bd215874c8]",
> "(ShardedThreadPool::shardedthreadpool_worker(unsigned int)+0x5c4)
> [0x55bd21c042a4]",
> "(ShardedThreadPool::WorkThreadSharded::entry()+0x14)
> [0x55bd21c07184]",
> "/lib64/libpthread.so.0(+0x81ca) [0x7f82e87c41ca]",
> "clone()"
> ],
> "ceph_version": "16.2.10",
> "crash_id":
> "2022-10-21T22:15:37.801458Z_ded9e80e-6c08-4502-8887-0438cdcd4a1c",
> "entity_name": "osd.21",
> "os_id": "centos",
> "os_name": "CentOS Stream",
> "os_version": "8",
> "os_version_id": "8",
> "process_name": "ceph-osd",
> "stack_sig":
> "a43cb3d3dcfcda8bffe97d30a1cb9c244ba20d595a2c8759a5cc8274781b0020",
> "timestamp": "2022-10-21T22:15:37.801458Z",
> "utsname_hostname": "node07",
> "utsname_machine": "x86_64",
> "utsname_release": "3.10.0-1160.45.1.el7.x86_64",
> "utsname_sysname": "Linux",
> "utsname_version": "#1 SMP Wed Oct 13 17:20:51 UTC 2021"
> }
>
> message info:
> [root@node07 ~]# dmesg -T | grep osd
> [Sat Oct 22 06:18:46 2022] tp_osd_tp[23817]: segfault at 55bd00000001 ip
> 00007f82e931cd6a sp 00007f82c6592f08 error 4 in
> libtcmalloc.so.4.5.3[7f82e92e1000+4d000]
> _______________________________________________
> ceph-users mailing list -- ceph-users@ceph.io
> To unsubscribe send an email to ceph-users-le...@ceph.io
There is currently a known race condition in onode reference counting
that affects all versions of Ceph [1][2]. Your backtrace is different
from everything else I've seen so far, though the top function being
Onode::put under a trimming operation is very suspicious. If you can
reproduce this and follow the tracker/report back after the fix, it
may be helpful to the community.
Cheers,
Tyler
[1] https://tracker.ceph.com/issues/57895
[2] https://tracker.ceph.com/issues/56382
_______________________________________________
ceph-users mailing list -- ceph-users@ceph.io
To unsubscribe send an email to ceph-users-le...@ceph.io
