Uploaded image for project: 'MariaDB Server'
  1. MariaDB Server
  2. MDEV-22460

safe_mutex: Found wrong usage of mutex 'LOCK_thd_data' and 'THR_LOCK_myisam', server with WSREP hangs

Details

    Description

      Note: It is filed separately from MDEV-22458 for the sake of a different mutex combination, specific to MyISAM. The test case has send/reap, and thus may be non-deterministic. The concurrency is added as it seems to be necessary for producing the mutex error message, hang is reproducible without it. I suppose send/reap can be removed from the test case for the regression suite.

      --source include/galera_cluster.inc
      		 
       
      		 
      CREATE TABLE t1 (a INT) ENGINE=MyISAM;
      		 
       
      		 
      --connect (con1,localhost,root,,test)
      		 
      --let $con1= `SELECT CONNECTION_ID()`
      		 
       
      		 
      --connection default
      		 
      --send
      		 
          INSERT INTO t1 VALUES (1),(2),(3),(4);
      		 
      --connection con1
      		 
          INSERT INTO t1 VALUES (5),(6),(7),(8);
      		 
      --connection default
      		 
      --reap
      		 
      --error ER_TARGET_NOT_EXPLAINABLE
      		 
      eval SHOW EXPLAIN FOR $con1;
      		 
      DROP TABLE t1;
      		 
      --connection con1
      		 
      SELECT 1;
      		 
       
      		 
      # Cleanup
      		 
      --disconnect con1

      Server hangs upon SELECT. The error log shows this:

      10.4 7f03a933

      		 
      safe_mutex: Found wrong usage of mutex 'LOCK_thd_data' and 'THR_LOCK_myisam'
      		 
      Mutex currently locked (in reverse order):
      		 
      THR_LOCK_myisam                   /data/src/10.4/storage/myisam/mi_close.c  line 35
      		 
      LOCK_thd_data                     /data/src/10.4/sql/sql_parse.cc  line 9017

      Reproducible on 10.4, 10.5.
      Not reproducible on 10.3.

      Attachments

        Issue Links

          relates to

          Bug - A problem which impairs or prevents the functions of the product. MDEV-22154 safe_mutex: Trying to lock mutex at <somewhere> , when the mutex was already locked at sql_parse.cc

          • Major - Major loss of function.
          • Closed

          Bug - A problem which impairs or prevents the functions of the product. MDEV-22458 Server with WSREP hangs after INSERT, wrong usage of mutex 'LOCK_thd_data' and 'share->intern_lock' / 'lock->mutex'

          • Major - Major loss of function.
          • Closed

          Activity

            ramesh Ramesh Sivaraman added a comment -

            jplindst Could not reproduce safe mutex wrong usage or wsrep SELECT hang issue on latest 10.4 build.

            10.4.23 12cad0c3468d734e041d4ef0cd5a26d2a28606fc (Debug)

            		 
            ==============================================================================
            		 
             
            		 
            TEST                                      RESULT   TIME (ms) or COMMENT
            		 
            --------------------------------------------------------------------------
            		 
             
            		 
            worker[1] Using MTR_BUILD_THREAD 300, with reserved ports 16000..16019
            		 
            connection node_2;
            		 
            connection node_1;
            		 
            CREATE TABLE t1 (a INT) ENGINE=MyISAM;
            		 
            connect  con1,localhost,root,,test;
            		 
            connection default;
            		 
            INSERT INTO t1 VALUES (1),(2),(3),(4);
            		 
            connection con1;
            		 
            INSERT INTO t1 VALUES (5),(6),(7),(8);
            		 
            connection default;
            		 
            SHOW EXPLAIN FOR 17;
            		 
            ERROR HY000: Target is not running an EXPLAINable command
            		 
            DROP TABLE t1;
            		 
            connection con1;
            		 
            SELECT 1;
            		 
            1
            		 
            1
            		 
            disconnect con1;
            		 
            galera.safemutex5 'innodb'               [ pass ]   2037
            		 
            --------------------------------------------------------------------------
            		 
            The servers were restarted 0 times
            		 
            Spent 2.037 of 20 seconds executing testcases
            		 
             
            		 
            Completed: All 1 tests were successful.

            ramesh Ramesh Sivaraman added a comment - jplindst Could not reproduce safe mutex wrong usage or wsrep SELECT hang issue on latest 10.4 build. 10.4.23 12cad0c3468d734e041d4ef0cd5a26d2a28606fc (Debug) ==============================================================================   TEST RESULT TIME (ms) or COMMENT --------------------------------------------------------------------------   worker[1] Using MTR_BUILD_THREAD 300, with reserved ports 16000..16019 connection node_2; connection node_1; CREATE TABLE t1 (a INT) ENGINE=MyISAM; connect con1,localhost,root,,test; connection default; INSERT INTO t1 VALUES (1),(2),(3),(4); connection con1; INSERT INTO t1 VALUES (5),(6),(7),(8); connection default; SHOW EXPLAIN FOR 17; ERROR HY000: Target is not running an EXPLAINable command DROP TABLE t1; connection con1; SELECT 1; 1 1 disconnect con1; galera.safemutex5 'innodb' [ pass ] 2037 -------------------------------------------------------------------------- The servers were restarted 0 times Spent 2.037 of 20 seconds executing testcases   Completed: All 1 tests were successful.

            People

              jplindst Jan Lindström (Inactive)
              elenst Elena Stepanova
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved:

                Git Integration

                  Error rendering 'com.xiplink.jira.git.jira_git_plugin:git-issue-webpanel'. Please contact your Jira administrators.