Fwd: OSD crash after change of osd_memory_target
Hi Igor, unfortunately same result: # ceph config dump WHO MASK LEVEL OPTION VALUE RO osd basic osd_memory_target 2147483648 # /usr/bin/ceph-osd -d --cluster ceph --id 0 --setuser ceph --setgroup ceph .... 0> 2020-01-23 10:48:04.436 7fc61b5b5c80 -1 *** Caught signal (Aborted) ** in thread 7fc61b5b5c80 thread_name:ceph-osd ceph version 14.2.6 (f0aa067ac7a02ee46ea48aa26c6e298b5ea272e9) nautilus (stable) 1: (()+0x12730) [0x7fc61c05d730] 2: (gsignal()+0x10b) [0x7fc61bb417bb] 3: (abort()+0x121) [0x7fc61bb2c535] 4: (()+0x8c983) [0x7fc61bef4983] 5: (()+0x928c6) [0x7fc61befa8c6] 6: (()+0x92901) [0x7fc61befa901] 7: (()+0x92b34) [0x7fc61befab34] 8: (()+0x5a3f53) [0x55ecdabb2f53] 9: (Option::size_t const md_config_t::get_val<Option::size_t>(ConfigValues const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const+0x81) [0x55ecdabb8c91] 10: (BlueStore::_set_cache_sizes()+0x15a) [0x55ecdb033d8a] 11: (BlueStore::_open_bdev(bool)+0x173) [0x55ecdb036b23] 12: (BlueStore::get_devices(std::set<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >*)+0xef) [0x55ecdb09d7ef] 13: (BlueStore::get_numa_node(int*, std::set<int, std::less<int>, std::allocator<int> >*, std::set<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >*)+0x7b) [0x55ecdb04571b] 14: (main()+0x2870) [0x55ecdab80440] 15: (__libc_start_main()+0xeb) [0x7fc61bb2e09b] 16: (_start()+0x2a) [0x55ecdabb2c6a] NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this. .... Best Regards, Martin Dne 22. 01. 20 v 16:33 Igor Fedotov napsal(a):
Hi Martin,
looks like a bug to me.
You might want to remove all custom settings from config database and try to set osd-memory-target only.
Would it help?
Thanks,
Igor
On 1/22/2020 3:43 PM, Martin Mlynář wrote:
Dne 21. 01. 20 v 21:12 Stefan Kooman napsal(a):
Quoting Martin Mlynář (nexus+ceph@smoula.net):
Do you think this could help? OSD does not even start, I'm getting a little lost how flushing caches could help. I might have mis-understood. I though the OSDs crashed when you set the config setting.
According to trace I suspect something around processing config values. I've just set the same config setting on a test cluster and restarted an OSD without problem. So, not sure what is going on there.
Gr. Stefan
I've compiled ceph-osd with debug symbols and got better backtrace:
-24> 2020-01-22 13:12:53.614 7f83ed064700 4 set_mon_vals no callback set -23> 2020-01-22 13:12:53.614 7f83ee867700 10 monclient: discarding stray monitor message auth_reply(proto 2 0 (0) Success) v1 -22> 2020-01-22 13:12:53.614 7f83ed064700 10 set_mon_vals osd_crush_update_on_start = true -21> 2020-01-22 13:12:53.614 7f83ed064700 10 set_mon_vals osd_max_backfills = 64 -20> 2020-01-22 13:12:53.614 7f83ed064700 10 set_mon_vals osd_memory_target = 2147483648 -19> 2020-01-22 13:12:53.614 7f83ed064700 10 set_mon_vals osd_recovery_max_active = 40 -18> 2020-01-22 13:12:53.614 7f83ed064700 10 set_mon_vals osd_recovery_max_single_start = 1000 -17> 2020-01-22 13:12:53.614 7f83ed064700 10 set_mon_vals osd_recovery_sleep_hdd = 0.000000 -16> 2020-01-22 13:12:53.614 7f83ed064700 10 set_mon_vals osd_recovery_sleep_hybrid = 0.000000 -15> 2020-01-22 13:12:53.627 7f83f0276c40 0 set uid:gid to 64045:64045 (ceph:ceph) -14> 2020-01-22 13:12:53.627 7f83f0276c40 0 ceph version 14.2.6 (f0aa067ac7a02ee46ea48aa26c6e298b5ea272e9) nautilus (stable), process ceph-osd, pid 1111622 -13> 2020-01-22 13:12:53.649 7f83f0276c40 0 pidfile_write: ignore empty --pid-file -12> 2020-01-22 13:12:53.657 7f83f0276c40 5 asok(0x5580518fa000) init /var/run/ceph/ceph-osd.6.asok -11> 2020-01-22 13:12:53.657 7f83f0276c40 5 asok(0x5580518fa000) bind_and_listen /var/run/ceph/ceph-osd.6.asok -10> 2020-01-22 13:12:53.657 7f83f0276c40 5 asok(0x5580518fa000) register_command 0 hook 0x558051872fc0 -9> 2020-01-22 13:12:53.657 7f83f0276c40 5 asok(0x5580518fa000) register_command version hook 0x558051872fc0 -8> 2020-01-22 13:12:53.657 7f83f0276c40 5 asok(0x5580518fa000) register_command git_version hook 0x558051872fc0 -7> 2020-01-22 13:12:53.657 7f83f0276c40 5 asok(0x5580518fa000) register_command help hook 0x558051874220 -6> 2020-01-22 13:12:53.657 7f83f0276c40 5 asok(0x5580518fa000) register_command get_command_descriptions hook 0x558051874260 -5> 2020-01-22 13:12:53.657 7f83ed865700 5 asok(0x5580518fa000) entry start -4> 2020-01-22 13:12:53.670 7f83f0276c40 5 object store type is bluestore -3> 2020-01-22 13:12:53.675 7f83f0276c40 1 bdev create path /var/lib/ceph/osd/ceph-6/block type kernel -2> 2020-01-22 13:12:53.675 7f83f0276c40 1 bdev(0x5580518f3f80 /var/lib/ceph/osd/ceph-6/block) open path /var/lib/ceph/osd/ceph-6/block -1> 2020-01-22 13:12:53.675 7f83f0276c40 1 bdev(0x5580518f3f80 /var/lib/ceph/osd/ceph-6/block) open size 3000588304384 (0x2baa1000000, 2.7 TiB) block_size 4096 (4 KiB) rotational discard not supported 0> 2020-01-22 13:12:53.714 7f83f0276c40 -1 *** Caught signal (Aborted) ** in thread 7f83f0276c40 thread_name:ceph-osd
ceph version 14.2.6 (f0aa067ac7a02ee46ea48aa26c6e298b5ea272e9) nautilus (stable) 1: (()+0x2c19654) [0x558045ec6654] 2: (()+0x12730) [0x7f83f0d1f730] 3: (gsignal()+0x10b) [0x7f83f08027bb] 4: (abort()+0x121) [0x7f83f07ed535] 5: (()+0x8c983) [0x7f83f0bb5983] 6: (()+0x928c6) [0x7f83f0bbb8c6] 7: (()+0x92901) [0x7f83f0bbb901] 8: (()+0x92b34) [0x7f83f0bbbb34] * 9: (void boost::throw_exception<boost::bad_get>(boost::bad_get const&)+0x7b) [0x5580454d5430]* * 10: (Option::size_t&& boost::relaxed_get<Option::size_t, boost::blank, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, unsigned long, long, double, bool, entity_addr_t, entity_addrvec_t, std::chrono::duration<long, std::ratio<1l, 1l> >, Option::size_t, uuid_d>(boost::variant<boost::blank, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, unsigned long, long, double, bool, entity_addr_t, entity_addrvec_t, std::chrono::duration<long, std::ratio<1l, 1l> >, Option::size_t, uuid_d>&&)+0x5b) [0x5580454d6223]* 11: (Option::size_t&& boost::strict_get<Option::size_t, boost::blank, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, unsigned long, long, double, bool, entity_addr_t, entity_addrvec_t, std::chrono::duration<long, std::ratio<1l, 1l> >, Option::size_t, uuid_d>(boost::variant<boost::blank, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, unsigned long, long, double, bool, entity_addr_t, entity_addrvec_t, std::chrono::duration<long, std::ratio<1l, 1l> >, Option::size_t, uuid_d>&&)+0x20) [0x5580454d4a39] 12: (Option::size_t&& boost::get<Option::size_t, boost::blank, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, unsigned long, long, double, bool, entity_addr_t, entity_addrvec_t, std::chrono::duration<long, std::ratio<1l, 1l> >, Option::size_t, uuid_d>(boost::variant<boost::blank, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, unsigned long, long, double, bool, entity_addr_t, entity_addrvec_t, std::chrono::duration<long, std::ratio<1l, 1l> >, Option::size_t, uuid_d>&&)+0x20) [0x5580454d1ed7] 13: (Option::size_t const md_config_t::get_val<Option::size_t>(ConfigValues const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const+0x48) [0x5580454ce882] 14: (Option::size_t const ConfigProxy::get_val<Option::size_t>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const+0x58) [0x5580454cb9b8] 15: (BlueStore::_set_cache_sizes()+0x159) [0x558045ce2213] 16: (BlueStore::_open_bdev(bool)+0x301) [0x558045ce6be3] 17: (BlueStore::get_devices(std::set<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >*)+0xf9) [0x558045d0f16d] 18: (BlueStore::get_numa_node(int*, std::set<int, std::less<int>, std::allocator<int> >*, std::set<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >*)+0x79) [0x558045d0eb55] 19: (main()+0x3aae) [0x5580454c2460] 20: (__libc_start_main()+0xeb) [0x7f83f07ef09b] 21: (_start()+0x2a) [0x5580454bda2a] NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.
And managed some gdb debugging (int BlueStore::_set_cache_sizes()):
(gdb) n 4116 cache_autotune_interval = (gdb) n 4117 cct->_conf.get_val<double>("bluestore_cache_autotune_interval"); (gdb) p cache_autotune_interval $3 = 5 (gdb) n 4118 osd_memory_target = cct->_conf.get_val<Option::size_t>("osd_memory_target"); (gdb) s std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string<std::allocator<char> > (this=0x7fffffffc140, __s=0x555558d26c2f "osd_memory_target", __a=...) at /usr/include/c++/8/bits/basic_string.h:515 515 : _M_dataplus(_M_local_data(), __a) (gdb) n 516 { _M_construct(__s, __s ? __s + traits_type::length(__s) : __s+npos); } (gdb) terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::bad_get>
' what(): boost::bad_get: failed value get using boost::get
But there I'm stuck. GDBing c++ code is a really dark sorcery for me.
Other get_vals look fine, maybe get_val<*Option::size_t*> is the problem? It looks like trouble outside of ceph - what system are you testing on? This is debian with official debian build from buster-backports. Maybe some of debian's patches?
-- Martin Mlynář
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
participants (1)
-
Martin Mlynář