[MDEV-26897] mysql server hang, Cannot connect Created: 2021-10-25  Updated: 2022-11-17  Resolved: 2022-11-17

Status: Closed
Project: MariaDB Server
Component/s: Storage Engine - TokuDB
Affects Version/s: 10.2.25
Fix Version/s: N/A

Type: Bug Priority: Major
Reporter: xie Assignee: Sergei Golubchik
Resolution: Won't Fix Votes: 0
Labels: TOKUDB_CHECKPOINT_LOCK, tokudb
Environment:

7.6.1810



 Description   

    191 __lll_lock_wait(libpthread.so.0),_L_lock_883(libpthread.so.0),pthread_mutex_lock(libpthread.so.0),je_choose_arena_hard(libjemalloc.so.1),calloc(libjemalloc.so.1),my_thread_init,init_new_connection_handler_thread,do_handle_one_connection,handle_one_connection,start_thread(libpthread.so.0),clone(libc.so.6)
      1 __lll_lock_wait(libpthread.so.0),_L_lock_883(libpthread.so.0),pthread_mutex_lock(libpthread.so.0),toku_txn_manager_clone_state_for_gc(ha_tokudb.so),txn_manager_state::init(ha_tokudb.so),toku_ftnode_leaf_run_gc(ha_tokudb.so),toku_ftnode_flush_callback(ha_tokudb.so),[clone(ha_tokudb.so),work_on_kibbutz(ha_tokudb.so),start_thread(libpthread.so.0),clone(libc.so.6)
      1 __lll_lock_wait(libpthread.so.0),_L_lock_883(libpthread.so.0),pthread_mutex_lock(libpthread.so.0),toku_txn_manager_clone_state_for_gc(ha_tokudb.so),txn_manager_state::init(ha_tokudb.so),toku_bnc_flush_to_child(ha_tokudb.so),toku_ft_flush_some_child(ha_tokudb.so),blocknum_s,(ha_tokudb.so),cleaner::run_cleaner(ha_tokudb.so),minicron_do(ha_tokudb.so),start_thread(libpthread.so.0),clone(libc.so.6)
      1 __lll_lock_wait(libpthread.so.0),_L_lock_883(libpthread.so.0),pthread_mutex_lock(libpthread.so.0),toku_mutex_lock_with_source_location(ha_tokudb.so),toku_txn_manager_start_txn(ha_tokudb.so),toku_txn_begin_with_xid(ha_tokudb.so),toku_txn_begin(ha_tokudb.so),ha_tokudb::write_metadata(ha_tokudb.so),ha_tokudb::write_to_status(ha_tokudb.so),ha_tokudb::update_max_auto_inc(ha_tokudb.so),ha_tokudb::end_bulk_insert(ha_tokudb.so),Write_rows_log_event::do_after_row_operations,Rows_log_event::do_apply_event,apply_event_and_update_pos_apply,handle_slave_sql,start_thread(libpthread.so.0),clone(libc.so.6)
      1 __lll_lock_wait(libpthread.so.0),_L_lock_883(libpthread.so.0),pthread_mutex_lock(libpthread.so.0),je_choose_arena_hard(libjemalloc.so.1),realloc(libjemalloc.so.1),pthread_getattr_np(libpthread.so.0),libmodule64::??(libmodule64.so),libmodule64::??(libmodule64.so),mmap_hook(libmodule64.so),pages_map.prop.2(libmodule64.so),je_chunk_alloc_mmap(libjemalloc.so.1),je_chunk_alloc(libjemalloc.so.1),je_base_alloc(libjemalloc.so.1),je_arenas_extend(libjemalloc.so.1),je_choose_arena_hard(libjemalloc.so.1),imemalign(libjemalloc.so.1),os_malloc_aligned(ha_tokudb.so),toku_xmalloc_aligned(ha_tokudb.so),toku_serialize_rollback_log_to(ha_tokudb.so),toku_rollback_flush_callback(ha_tokudb.so),[clone(ha_tokudb.so),work_on_kibbutz(ha_tokudb.so),start_thread(libpthread.so.0),clone(libc.so.6)
      1 __lll_lock_wait(libpthread.so.0),_L_lock_883(libpthread.so.0),pthread_mutex_lock(libpthread.so.0),je_choose_arena_hard(libjemalloc.so.1),malloc(libjemalloc.so.1),toku_xmalloc(ha_tokudb.so),toku_txn_manager_clone_state_for_gc(ha_tokudb.so),txn_manager_state::init(ha_tokudb.so),toku_ftnode_leaf_run_gc(ha_tokudb.so),toku_ftnode_flush_callback(ha_tokudb.so),[clone(ha_tokudb.so),work_on_kibbutz(ha_tokudb.so),start_thread(libpthread.so.0),clone(libc.so.6)
      1 __lll_lock_wait(libpthread.so.0),_L_lock_883(libpthread.so.0),pthread_mutex_lock(libpthread.so.0),je_base_node_alloc(libjemalloc.so.1),je_huge_palloc(libjemalloc.so.1),malloc(libjemalloc.so.1),my_malloc,hp_get_new_block,heap_write,ha_heap::write_row,handler::ha_write_tmp_row,schema_table_store_record,show_status_array,fill_variables,get_schema_tables_result,JOIN::exec_inner,JOIN::exec,mysql_select,handle_select,execute_sqlcom_select,mysql_execute_command,mysql_parse,dispatch_command,do_command,do_handle_one_connection,handle_one_connection,start_thread(libpthread.so.0),clone(libc.so.6)
      1 __lll_lock_wait(libpthread.so.0),_L_lock_883(libpthread.so.0),pthread_mutex_lock(libpthread.so.0),je_base_node_alloc(libjemalloc.so.1),je_chunk_unmap(libjemalloc.so.1),je_huge_dalloc(libjemalloc.so.1),toku_mempool_destroy(ha_tokudb.so),bn_data::destroy(ha_tokudb.so),destroy_basement_node(ha_tokudb.so),toku_ftnode_leaf_rebalance(ha_tokudb.so),void**,(ha_tokudb.so),checkpointer::checkpoint_pending_pairs(ha_tokudb.so),checkpointer::end_checkpoint(ha_tokudb.so),toku_checkpoint(ha_tokudb.so),[clone(ha_tokudb.so),minicron_do(ha_tokudb.so),start_thread(libpthread.so.0),clone(libc.so.6)



 Comments   
Comment by Sergei Golubchik [ 2021-10-25 ]

I presume, it's the output from pmp

Comment by xie [ 2021-10-26 ]

@Sergei Golubchik
Yes, I analyzed it by myself.
I think it's the reason for jemalloc. I hope the bosses can help me confirm and fix this problem

Comment by xie [ 2021-10-26 ]

libmodule64 ,maybe this https://knowledge.informatica.com/s/article/Azul-Java-process-hangs-during-native-library-load?language=en_US

Comment by renaesop [ 2022-11-14 ]

look like cortex trap bug。

Comment by Sergei Golubchik [ 2022-11-17 ]

10.2 is out of support.
TokuDB was disabled in MariaDB and removed in 10.6

Generated at Thu Feb 08 09:48:47 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.