Hi all,
Few days after an Octopus to Pacific upgrade, I have a crashed Ceph Cluster.
Most of the OSD are down (6 on 8) and crash on start.
Seems like a lot to https://forum.proxmox.com/threads/ceph-16-2-pacific-cluster-crash.92367/ but switch bluestore_allocator an bluefs_allocator to bitmap mode do not help.
Truncated (too long to be posted) log output from a crashed OSD below.
Any advice?
Few days after an Octopus to Pacific upgrade, I have a crashed Ceph Cluster.
Most of the OSD are down (6 on 8) and crash on start.
Seems like a lot to https://forum.proxmox.com/threads/ceph-16-2-pacific-cluster-crash.92367/ but switch bluestore_allocator an bluefs_allocator to bitmap mode do not help.
Truncated (too long to be posted) log output from a crashed OSD below.
Any advice?
Code:
janv. 06 14:53:08 pve11 systemd[1]: Starting Ceph object storage daemon osd.0...
janv. 06 14:53:08 pve11 systemd[1]: Started Ceph object storage daemon osd.0.
janv. 06 14:53:19 pve11 ceph-osd[24802]: 2022-01-06T14:53:19.214+0100 7f2d01c05f00 -1 bluefs _allocate allocation failed, needed 0x8025e
janv. 06 14:53:19 pve11 ceph-osd[24802]: 2022-01-06T14:53:19.214+0100 7f2d01c05f00 -1 bluefs _flush_range allocated: 0x0 offset: 0x0 length: 0x8025e
janv. 06 14:53:19 pve11 ceph-osd[24802]: ./src/os/bluestore/BlueFS.cc: In function 'int BlueFS::_flush_range(BlueFS::FileWriter*, uint64_t, uint64_t)' thread 7f2d01c05f00 time 2022-01-06T14:53:19.219438+0100
janv. 06 14:53:19 pve11 ceph-osd[24802]: ./src/os/bluestore/BlueFS.cc: 2768: ceph_abort_msg("bluefs enospc")
janv. 06 14:53:19 pve11 ceph-osd[24802]: ceph version 16.2.7 (f9aa029788115b5df5eeee328f584156565ee5b7) pacific (stable)
janv. 06 14:53:19 pve11 ceph-osd[24802]: 1: (ceph::__ceph_abort(char const*, int, char const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)+0xd3) [0x564c8796f0df]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 2: (BlueFS::_flush_range(BlueFS::FileWriter*, unsigned long, unsigned long)+0x9bd) [0x564c88057bbd]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 3: (BlueFS::_flush(BlueFS::FileWriter*, bool, bool*)+0x9a) [0x564c880581ca]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 4: (BlueFS::_flush(BlueFS::FileWriter*, bool, std::unique_lock<std::mutex>&)+0x2f) [0x564c8806945f]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 5: (BlueRocksWritableFile::Append(rocksdb::Slice const&)+0x100) [0x564c880817d0]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 6: (rocksdb::LegacyWritableFileWrapper::Append(rocksdb::Slice const&, rocksdb::IOOptions const&, rocksdb::IODebugContext*)+0x48) [0x564c8854824e]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 7: (rocksdb::WritableFileWriter::WriteBuffered(char const*, unsigned long)+0x338) [0x564c88722d18]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 8: (rocksdb::WritableFileWriter::Append(rocksdb::Slice const&)+0x5d7) [0x564c8872129b]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 9: (rocksdb::BlockBasedTableBuilder::WriteRawBlock(rocksdb::Slice const&, rocksdb::CompressionType, rocksdb::BlockHandle*, bool)+0x11d) [0x564c888eb2d7]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 10: (rocksdb::BlockBasedTableBuilder::WriteBlock(rocksdb::Slice const&, rocksdb::BlockHandle*, bool)+0x7d0) [0x564c888eb0be]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 11: (rocksdb::BlockBasedTableBuilder::WriteBlock(rocksdb::BlockBuilder*, rocksdb::BlockHandle*, bool)+0x48) [0x564c888ea8da]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 12: (rocksdb::BlockBasedTableBuilder::Flush()+0x9a) [0x564c888ea88a]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 13: (rocksdb::BlockBasedTableBuilder::Add(rocksdb::Slice const&, rocksdb::Slice const&)+0x197) [0x564c888ea3bf]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 14: (rocksdb::BuildTable(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, rocksdb::Env*, rocksdb::FileSystem*, rocksdb::ImmutableCFOptions const&, rocksdb::MutableCFOptions const&, rocksdb::FileOptions const&, rocksdb::TableCache*, ro>
janv. 06 14:53:19 pve11 ceph-osd[24802]: 15: (rocksdb::DBImpl::WriteLevel0TableForRecovery(int, rocksdb::ColumnFamilyData*, rocksdb::MemTable*, rocksdb::VersionEdit*)+0x5ea) [0x564c885e6226]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 16: (rocksdb::DBImpl::RecoverLogFiles(std::vector<unsigned long, std::allocator<unsigned long> > const&, unsigned long*, bool, bool*)+0x1ad1) [0x564c885e4e9d]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 17: (rocksdb::DBImpl::Recover(std::vector<rocksdb::ColumnFamilyDescriptor, std::allocator<rocksdb::ColumnFamilyDescriptor> > const&, bool, bool, bool, unsigned long*)+0x159e) [0x564c885e23d4]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 18: (rocksdb::DBImpl::Open(rocksdb::DBOptions const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::vector<rocksdb::ColumnFamilyDescriptor, std::allocator<rocksdb::ColumnFamilyDescriptor> > const&, std::vector<rocksdb::ColumnF>
janv. 06 14:53:19 pve11 ceph-osd[24802]: 19: (rocksdb::DB::Open(rocksdb::DBOptions const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::vector<rocksdb::ColumnFamilyDescriptor, std::allocator<rocksdb::ColumnFamilyDescriptor> > const&, std::vector<rocksdb::ColumnFamil>
janv. 06 14:53:19 pve11 ceph-osd[24802]: 20: (RocksDBStore::do_open(std::ostream&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)+0x10a6) [0x564c884f78b6]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 21: (BlueStore::_open_db(bool, bool, bool)+0xa19) [0x564c87f75b19]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 22: (BlueStore::_open_db_and_around(bool, bool)+0x332) [0x564c87fbab92]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 23: (BlueStore::_mount()+0x191) [0x564c87fbd531]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 24: (OSD::init()+0x58d) [0x564c87a645ed]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 25: main()
janv. 06 14:53:19 pve11 ceph-osd[24802]: 26: __libc_start_main()
janv. 06 14:53:19 pve11 ceph-osd[24802]: 27: _start()
janv. 06 14:53:19 pve11 ceph-osd[24802]: *** Caught signal (Aborted) **
janv. 06 14:53:19 pve11 ceph-osd[24802]: in thread 7f2d01c05f00 thread_name:ceph-osd
janv. 06 14:53:19 pve11 ceph-osd[24802]: 2022-01-06T14:53:19.234+0100 7f2d01c05f00 -1 ./src/os/bluestore/BlueFS.cc: In function 'int BlueFS::_flush_range(BlueFS::FileWriter*, uint64_t, uint64_t)' thread 7f2d01c05f00 time 2022-01-06T14:53:19.219438+0100
janv. 06 14:53:19 pve11 ceph-osd[24802]: ./src/os/bluestore/BlueFS.cc: 2768: ceph_abort_msg("bluefs enospc")
janv. 06 14:53:19 pve11 ceph-osd[24802]: ceph version 16.2.7 (f9aa029788115b5df5eeee328f584156565ee5b7) pacific (stable)
janv. 06 14:53:19 pve11 ceph-osd[24802]: 1: (ceph::__ceph_abort(char const*, int, char const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)+0xd3) [0x564c8796f0df]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 2: (BlueFS::_flush_range(BlueFS::FileWriter*, unsigned long, unsigned long)+0x9bd) [0x564c88057bbd]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 3: (BlueFS::_flush(BlueFS::FileWriter*, bool, bool*)+0x9a) [0x564c880581ca]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 4: (BlueFS::_flush(BlueFS::FileWriter*, bool, std::unique_lock<std::mutex>&)+0x2f) [0x564c8806945f]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 5: (BlueRocksWritableFile::Append(rocksdb::Slice const&)+0x100) [0x564c880817d0]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 6: (rocksdb::LegacyWritableFileWrapper::Append(rocksdb::Slice const&, rocksdb::IOOptions const&, rocksdb::IODebugContext*)+0x48) [0x564c8854824e]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 7: (rocksdb::WritableFileWriter::WriteBuffered(char const*, unsigned long)+0x338) [0x564c88722d18]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 8: (rocksdb::WritableFileWriter::Append(rocksdb::Slice const&)+0x5d7) [0x564c8872129b]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 9: (rocksdb::BlockBasedTableBuilder::WriteRawBlock(rocksdb::Slice const&, rocksdb::CompressionType, rocksdb::BlockHandle*, bool)+0x11d) [0x564c888eb2d7]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 10: (rocksdb::BlockBasedTableBuilder::WriteBlock(rocksdb::Slice const&, rocksdb::BlockHandle*, bool)+0x7d0) [0x564c888eb0be]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 11: (rocksdb::BlockBasedTableBuilder::WriteBlock(rocksdb::BlockBuilder*, rocksdb::BlockHandle*, bool)+0x48) [0x564c888ea8da]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 12: (rocksdb::BlockBasedTableBuilder::Flush()+0x9a) [0x564c888ea88a]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 13: (rocksdb::BlockBasedTableBuilder::Add(rocksdb::Slice const&, rocksdb::Slice const&)+0x197) [0x564c888ea3bf]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 14: (rocksdb::BuildTable(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, rocksdb::Env*, rocksdb::FileSystem*, rocksdb::ImmutableCFOptions const&, rocksdb::MutableCFOptions const&, rocksdb::FileOptions const&, rocksdb::TableCache*, ro>
janv. 06 14:53:19 pve11 ceph-osd[24802]: 15: (rocksdb::DBImpl::WriteLevel0TableForRecovery(int, rocksdb::ColumnFamilyData*, rocksdb::MemTable*, rocksdb::VersionEdit*)+0x5ea) [0x564c885e6226]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 16: (rocksdb::DBImpl::RecoverLogFiles(std::vector<unsigned long, std::allocator<unsigned long> > const&, unsigned long*, bool, bool*)+0x1ad1) [0x564c885e4e9d]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 17: (rocksdb::DBImpl::Recover(std::vector<rocksdb::ColumnFamilyDescriptor, std::allocator<rocksdb::ColumnFamilyDescriptor> > const&, bool, bool, bool, unsigned long*)+0x159e) [0x564c885e23d4]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 18: (rocksdb::DBImpl::Open(rocksdb::DBOptions const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::vector<rocksdb::ColumnFamilyDescriptor, std::allocator<rocksdb::ColumnFamilyDescriptor> > const&, std::vector<rocksdb::ColumnF>
janv. 06 14:53:19 pve11 ceph-osd[24802]: 19: (rocksdb::DB::Open(rocksdb::DBOptions const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::vector<rocksdb::ColumnFamilyDescriptor, std::allocator<rocksdb::ColumnFamilyDescriptor> > const&, std::vector<rocksdb::ColumnFamil>
janv. 06 14:53:19 pve11 ceph-osd[24802]: 20: (RocksDBStore::do_open(std::ostream&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)+0x10a6) [0x564c884f78b6]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 21: (BlueStore::_open_db(bool, bool, bool)+0xa19) [0x564c87f75b19]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 22: (BlueStore::_open_db_and_around(bool, bool)+0x332) [0x564c87fbab92]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 23: (BlueStore::_mount()+0x191) [0x564c87fbd531]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 24: (OSD::init()+0x58d) [0x564c87a645ed]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 25: main()
janv. 06 14:53:19 pve11 ceph-osd[24802]: 26: __libc_start_main()
janv. 06 14:53:19 pve11 ceph-osd[24802]: 27: _start()
janv. 06 14:53:19 pve11 ceph-osd[24802]: ceph version 16.2.7 (f9aa029788115b5df5eeee328f584156565ee5b7) pacific (stable)
janv. 06 14:53:19 pve11 ceph-osd[24802]: 1: /lib/x86_64-linux-gnu/libpthread.so.0(+0x14140) [0x7f2d0225d140]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 2: gsignal()
janv. 06 14:53:19 pve11 ceph-osd[24802]: 3: abort()
janv. 06 14:53:19 pve11 ceph-osd[24802]: 4: (ceph::__ceph_abort(char const*, int, char const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)+0x18a) [0x564c8796f196]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 5: (BlueFS::_flush_range(BlueFS::FileWriter*, unsigned long, unsigned long)+0x9bd) [0x564c88057bbd]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 6: (BlueFS::_flush(BlueFS::FileWriter*, bool, bool*)+0x9a) [0x564c880581ca]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 7: (BlueFS::_flush(BlueFS::FileWriter*, bool, std::unique_lock<std::mutex>&)+0x2f) [0x564c8806945f]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 8: (BlueRocksWritableFile::Append(rocksdb::Slice const&)+0x100) [0x564c880817d0]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 9: (rocksdb::LegacyWritableFileWrapper::Append(rocksdb::Slice const&, rocksdb::IOOptions const&, rocksdb::IODebugContext*)+0x48) [0x564c8854824e]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 10: (rocksdb::WritableFileWriter::WriteBuffered(char const*, unsigned long)+0x338) [0x564c88722d18]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 11: (rocksdb::WritableFileWriter::Append(rocksdb::Slice const&)+0x5d7) [0x564c8872129b]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 12: (rocksdb::BlockBasedTableBuilder::WriteRawBlock(rocksdb::Slice const&, rocksdb::CompressionType, rocksdb::BlockHandle*, bool)+0x11d) [0x564c888eb2d7]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 13: (rocksdb::BlockBasedTableBuilder::WriteBlock(rocksdb::Slice const&, rocksdb::BlockHandle*, bool)+0x7d0) [0x564c888eb0be]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 14: (rocksdb::BlockBasedTableBuilder::WriteBlock(rocksdb::BlockBuilder*, rocksdb::BlockHandle*, bool)+0x48) [0x564c888ea8da]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 15: (rocksdb::BlockBasedTableBuilder::Flush()+0x9a) [0x564c888ea88a]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 16: (rocksdb::BlockBasedTableBuilder::Add(rocksdb::Slice const&, rocksdb::Slice const&)+0x197) [0x564c888ea3bf]
janv. 06 14:53:19 pve11 ceph-osd[24802]: 17: (rocksdb::BuildTable(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, rocksdb::Env*, rocksdb::FileSystem*, rocksdb::ImmutableCFOptions const&, rocksdb::MutableCFOptions const&, rocksdb::FileOptions const&, rocksdb::TableCache*, ro>
janv. 06 14:53:19 pve11 ceph-osd[24802]: 18: (rocksdb::DBImpl::WriteLevel0TableForRecovery(int, rocksdb::ColumnFamilyData*, rocksdb::MemTable*, rocksdb::VersionEdit*)+0x5ea) [0x564c885e6226]
...