Uploaded image for project: 'MariaDB Server'
  1. MariaDB Server
  2. MDEV-24307

Crashes in ALTER TABLE IMPORT TABLESPACE, Assertion dict0dict.cc, table->get_ref_count() == 0

    XMLWordPrintable

    Details

      Description

      I am daily replicating some tables between two machines using .ibd file transfer. I randomly get crashes when doing this and i replicate one schema at a time.

      On source machine:

      • First using FLUSH TABLES ... FOR EXPORT;
      • Copy the .ibd files
      • UNLOCK TABLES;

      On target:

      • LOCK TABLES ... LOW_PRIORITY WRITE;

      On target for each table:

      • ALTER TABLE <schema>.<tbl> DISCARD TABLESPACE;
      • Copy ibd file in place then set proper permissions
      • ALTER TABLE xxx IMPORT TABLESPACE. ( <== Here is where it always break, and on varying tables)

      On target:

      • UNLOCK TABLES;

      Example log below, i tried to follow the procedure by docs but might be overlooking something? I do things like dropping indexes on large table at source when reloading to get speed but this arises also on tables where i do not do this so do not see it would be related to a specific table modification at the source. Both machines are running same version and same OS.

      My solution now is simply make another try after db service restarts itself and then it seem to work. It keeps me afloat but it's a really ugly "solution" as i have plenty of services connected to this provider machine. I want to have "hotswap" of data all data at once and only at a specific trigger once a day when its ready cooked in source and that's why i'am using this approach instead of using a galera cluster setup.

      Thank you so much in advance for having look into this. Let me know what more data i could provide to ease the review.

      Best Regards
      /Daniel

      From log
      ======
      2020-11-30 09:08:51 0x7fc8644e0700 InnoDB: Assertion failure in file /home/buildbot/buildbot/padding_for_CPACK_RPM_BUILD_SOURCE_DIRS_PREFIX/mariadb-10.5.5/storage/innobase/dict/dict0dict.cc line 1918
      InnoDB: Failing assertion: table->get_ref_count() == 0
      InnoDB: We intentionally generate a memory trap.
      InnoDB: Submit a detailed bug report to https://jira.mariadb.org/
      InnoDB: If you get repeated assertion failures or crashes, even
      InnoDB: immediately after the mysqld startup, there may be
      InnoDB: corruption in the InnoDB tablespace. Please refer to
      InnoDB: https://mariadb.com/kb/en/library/innodb-recovery-modes/
      InnoDB: about forcing recovery.
      201130 9:08:51 [ERROR] mysqld got signal 6 ;
      This could be because you hit a bug. It is also possible that this binary
      or one of the libraries it was linked against is corrupt, improperly built,
      or misconfigured. This error can also be caused by malfunctioning hardware.

      To report this bug, see https://mariadb.com/kb/en/reporting-bugs

      We will try our best to scrape up some info that will hopefully help
      diagnose the problem, but since we have already crashed,
      something is definitely wrong and this may fail.

      Server version: 10.5.5-MariaDB
      key_buffer_size=20971520
      read_buffer_size=131072
      max_used_connections=7
      max_threads=153
      thread_count=7
      It is possible that mysqld could use up to
      key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 357211 K bytes of memory
      Hope that's ok; if not, decrease some variables in the equation.

      Thread pointer: 0x7fc064000c58
      Attempting backtrace. You can use the following information to find out
      where mysqld died. If you see no messages after this, something went
      terribly wrong...
      stack_bottom = 0x7fc8644dfbd8 thread_stack 0x49000
      ??:0(my_print_stacktrace)[0x55a3ca399dee]
      ??:0(handle_fatal_signal)[0x55a3c9e24ec5]
      sigaction.c:0(__restore_rt)[0x7fc8762dedd0]
      :0(__GI_raise)[0x7fc87407d70f]
      :0(__GI_abort)[0x7fc874067b25]
      ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a3c9b0ac23]
      ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a3c9b18b1c]
      ??:0(wsrep_notify_status(wsrep::server_state::state, wsrep::view const*))[0x55a3ca0f29fa]
      ??:0(mysql_discard_or_import_tablespace(THD*, TABLE_LIST*, bool))[0x55a3c9cc4f77]
      ??:0(Sql_cmd_discard_import_tablespace::execute(THD*))[0x55a3c9d2b02d]
      ??:0(mysql_execute_command(THD*))[0x55a3c9c38fed]
      ??:0(mysql_parse(THD*, char*, unsigned int, Parser_state*, bool, bool))[0x55a3c9c2bc62]
      ??:0(dispatch_command(enum_server_command, THD*, char*, unsigned int, bool, bool))[0x55a3c9c360fe]
      ??:0(do_command(THD*))[0x55a3c9c3724f]
      ??:0(do_handle_one_connection(CONNECT*, bool))[0x55a3c9d26bc4]
      ??:0(handle_one_connection)[0x55a3c9d26f7d]
      ??:0(MyCTX_nopad::finish(unsigned char*, unsigned int*))[0x55a3ca043d5a]
      pthread_create.c:0(start_thread)[0x7fc8762d42de]
      :0(_GI__clone)[0x7fc874141e83]

      Trying to get some variables.
      Some pointers may be invalid and cause the dump to abort.
      Query (0x7fc0640109d0): ALTER TABLE `dwf`.`billing` IMPORT TABLESPACE
      Connection ID (thread ID): 4
      Status: NOT_KILLED

      Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=on,table_elimination=on,extended_keys=on,exists_to_in=on,orderby_uses_equalities=on,condition_pushdown_for_derived=on,split_materialized=on,condition_pushdown_for_subquery=on,rowid_filter=on,condition_pushdown_from_having=on,not_null_range_scan=off

      The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains
      information that should help you find out what is causing the crash.
      Writing a core file...
      Working directory at /b001/dwfdb-data/mysql
      Resource Limits:
      Limit Soft Limit Hard Limit Units
      Max cpu time unlimited unlimited seconds
      Max file size unlimited unlimited bytes
      Max data size unlimited unlimited bytes
      Max stack size 8388608 unlimited bytes
      Max core file size unlimited unlimited bytes
      Max resident set unlimited unlimited bytes
      Max processes 192044 192044 processes
      Max open files 16384 16384 files
      Max locked memory 65536 65536 bytes
      Max address space unlimited unlimited bytes
      Max file locks unlimited unlimited locks
      Max pending signals 192044 192044 signals
      Max msgqueue size 819200 819200 bytes
      Max nice priority 0 0
      Max realtime priority 0 0
      Max realtime timeout unlimited unlimited us
      Core pattern: |/usr/lib/systemd/systemd-coredump %P %u %g %s %t %c %h %e

        Attachments

        1. case1_gdb.txt
          64 kB
        2. case1_log.log
          10 kB
        3. case2_gdb.txt
          26 kB
        4. case2_log.log
          4 kB
        5. mdev24307_py_mariadb_reproduce.py
          7 kB
        6. mdev24307_reproduce.py
          7 kB

          Activity

            People

            Assignee:
            marko Marko Mäkelä
            Reporter:
            dene Daniel Nilsson
            Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

              Dates

              Created:
              Updated:

                Git Integration