Uploaded image for project: 'MariaDB Server'
  1. MariaDB Server
  2. MDEV-26550

[FATAL] [ERROR] mysqld got signal 6 ;InnoDB: Aborting because of a corrupt database page.

Details

    Description

      when version up from 10.1.12 to 10.3.8 , I got error on mariadb, following error occured and cannot mysqldump on mariadb. I want to know if there is a patch version related to the error below.

      Please refer to the error log below.

      [In case of failure, all logs are in the attached file.]

      374116 2021-08-25 18:47:54 0 [Note] InnoDB: Rollback of non-prepared transactions completed
      374117 2021-08-25 18:47:54 0 [Note] InnoDB: Removed temporary tablespace data file: "ibtmp1"
      374118 2021-08-25 18:47:54 0 [Note] InnoDB: Creating shared tablespace for temporary tables
      374119 2021-08-25 18:47:54 0 [Note] InnoDB: Setting file '/hiware6/dbms/data/datadir/ibtmp1' size to 12 MB. Physically writing the file full; Pleasewait ...
      374120 2021-08-25 18:47:54 0 [Note] InnoDB: File '/hiware6/dbms/data/datadir/ibtmp1' size is now 12 MB.
      374121 2021-08-25 18:47:54 0 [Note] InnoDB: Waiting for purge to start
      374122 2021-08-25 18:47:54 0 [Note] InnoDB: 10.3.8 started; log sequence number 18782816143728; transaction id 20058497144
      374123 2021-08-25 18:47:54 0 [Note] Plugin 'FEEDBACK' is disabled.
      374124 2021-08-25 18:47:54 0 [Note] Recovering after a crash using /hiware6/dbms/log/binary/mysql-bin
      374125 2021-08-25 18:47:54 0 [Note] Starting crash recovery...
      374126 2021-08-25 18:47:54 0 [Note] InnoDB: Starting recovery for XA transactions...
      374127 2021-08-25 18:47:54 0 [Note] InnoDB: Transaction 20058497141 in prepared state after recovery
      374128 2021-08-25 18:47:54 0 [Note] InnoDB: Transaction contains changes to 1 rows
      374129 2021-08-25 18:47:54 0 [Note] InnoDB: 1 transactions in prepared state after recovery
      374130 2021-08-25 18:47:54 0 [Note] Found 1 prepared transaction(s) in InnoDB
      374131 2021-08-25 18:47:54 0 [Note] Crash recovery finished.
      374132 2021-08-25 18:47:54 0 [Note] Server socket created on IP: '::'.
      374133 2021-08-25 18:47:54 0 [Warning] 'proxies_priv' entry '@% root@hssecgw1' ignored in --skip-name-resolve mode.
      374134 2021-08-25 18:47:54 0 [Note] Reading of all Master_info entries succeded
      374135 2021-08-25 18:47:54 0 [Note] Added new Master_info '' to hash table
      374136 2021-08-25 18:47:54 0 [Note] /hiware6/dbms/engine/mariadb/bin/mysqld: ready for connections.
      374137 Version: '10.3.8-MariaDB-log' socket: '/hiware6/dbms/data/mysql.sock' port: 25317 MariaDB Server
      374138 2021-08-25 18:47:55 1 [ERROR] [FATAL] InnoDB: Aborting because of a corrupt database page.
      374139 210825 18:47:55 [ERROR] mysqld got signal 6 ;
      374140 This could be because you hit a bug. It is also possible that this binary
      374141 or one of the libraries it was linked against is corrupt, improperly built,
      374142 or misconfigured. This error can also be caused by malfunctioning hardware.
      374143
      374144 To report this bug, see https://mariadb.com/kb/en/reporting-bugs
      374145
      374146 We will try our best to scrape up some info that will hopefully help
      374147 diagnose the problem, but since we have already crashed,
      374148 something is definitely wrong and this may fail.
      374149
      374150 Server version: 10.3.8-MariaDB-log
      374151 key_buffer_size=134217728
      374152 read_buffer_size=2097152
      374153 max_used_connections=1
      374154 max_threads=65537
      374155 thread_count=7
      374156 It is possible that mysqld could use up to
      374157 key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 201527789 K bytes of memory
      374158 Hope that's ok; if not, decrease some variables in the equation.
      374159
      374160 Thread pointer: 0x7fbf380009a8
      374161 Attempting backtrace. You can use the following information to find out
      374162 where mysqld died. If you see no messages after this, something went
      374163 terribly wrong...
      374164 stack_bottom = 0x7fbf4ef9ed08 thread_stack 0x49000
      374165 mysys/stacktrace.c:270(my_print_stacktrace)[0x55a5153ddb1b]
      374166 sql/signal_handler.cc:168(handle_fatal_signal)[0x55a514eca7c7]

      374165 mysys/stacktrace.c:270(my_print_stacktrace)[0x55a5153ddb1b]
      374166 sql/signal_handler.cc:168(handle_fatal_signal)[0x55a514eca7c7]
      374167 sigaction.c:0(__restore_rt)[0x7fc81a0435e0]
      374168 :0(__GI_raise)[0x7fc818e571f7]
      374169 :0(__GI_abort)[0x7fc818e588e8]
      374170 ut/ut0ut.cc:767(ib::fatal::~fatal())[0x55a5151b799a]
      374171 buf/buf0buf.cc:6226(buf_page_io_complete(buf_page_t*, bool, bool))[0x55a5151f8e11]
      374172 buf/buf0rea.cc:227(buf_read_page_low(dberr_t*, bool, unsigned long, unsigned long, page_id_t const&, page_size_t const&, bool, bool))[0x55a515217170]
      374173 :0(buf_read_page(page_id_t const&, page_size_t const&))[0x55a5152184d6]
      374174 :0(buf_page_get_gen(page_id_t const&, page_size_t const&, unsigned long, buf_block_t*, unsigned long, char const*, unsigned int, mtr_t*, dberr_t*))[0x55a5151f6023]
      374175 fsp/fsp0fsp.cc:451(fseg_free_step_func(unsigned char*, bool, mtr_t*))[0x55a51527b1d7]
      374176 trx/trx0purge.cc:402(trx_purge_truncate_rseg_history(trx_rseg_t&, purge_sys_t::iterator const&))[0x55a515180b11]
      374177 :0(trx_purge(unsigned long, bool))[0x55a515187621]
      374178 srv/srv0srv.cc:2585(srv_purge_coordinator_thread)[0x55a51516b63c]
      374179 pthread_create.c:0(start_thread)[0x7fc81a03be25]
      374180 /lib64/libc.so.6(clone+0x6d)[0x7fc818f1a34d]
      374181
      374182 Trying to get some variables.
      374183 Some pointers may be invalid and cause the dump to abort.
      374184 Query (0x0):
      374185 Connection ID (thread ID): 1
      374186 Status: NOT_KILLED
      374187
      374188 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=off,table_elimination=on,extended_keys=on,exists_to_in=on,orderby_uses_equalities=on,condition_pushdown_for_derived=on,split_materialized=on
      374189
      374190 The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains
      374191 information that should help you find out what is causing the crash.
      374192 2021-08-25 18:47:56 0 [Note] Using unique option prefix 'myisam_recover' is error-prone and can break in the future. Please use the full name'myisam-recover-options' instead.
      374193 2021-08-25 18:47:56 0 [Note] InnoDB: Using Linux native AIO
      374194 2021-08-25 18:47:56 0 [Note] InnoDB: Mutexes and rw_locks use GCC atomic builtins
      374195 2021-08-25 18:47:56 0 [Note] InnoDB: Uses event mutexes
      374196 2021-08-25 18:47:56 0 [Note] InnoDB: Compressed tables use zlib 1.2.3
      374197 2021-08-25 18:47:56 0 [Note] InnoDB: Number of pools: 1
      374198 2021-08-25 18:47:56 0 [Note] InnoDB: Using SSE2 crc32 instructions
      374199 2021-08-25 18:47:56 0 [Note] InnoDB: Initializing buffer pool, total size = 32G, instances = 8, chunk size = 128M
      374200 2021-08-25 18:47:57 0 [Note] InnoDB: Completed initialization of buffer pool
      374201 2021-08-25 18:47:57 0 [Note] InnoDB: If the mysqld execution user is authorized, page cleaner thread priority can be changed. See the man pageof setpriority().
      374202 2021-08-25 18:47:57 0 [Note] InnoDB: Starting crash recovery from checkpoint LSN=18782811967559

      Attachments

        Issue Links

          Activity

            I do not think that we have fixed the error handling in all code paths. MDEV-13542 has been filed for this problem.

            I think that some error reporting should have been improved. In the 10.5 release, I would expect us to report the file name, but we could still abort the server in this case.

            marko Marko Mäkelä added a comment - I do not think that we have fixed the error handling in all code paths. MDEV-13542 has been filed for this problem. I think that some error reporting should have been improved. In the 10.5 release, I would expect us to report the file name, but we could still abort the server in this case.
            Sangmin Sangmin Han added a comment - - edited

            Thanks for the comment. What you mean is, it is speculated that the problem is caused by corruption of a specific file.

            Is it correct to assume that the problem is caused by corruption of a specific file?

            Sangmin Sangmin Han added a comment - - edited Thanks for the comment. What you mean is, it is speculated that the problem is caused by corruption of a specific file. Is it correct to assume that the problem is caused by corruption of a specific file?

            Yes, a particular page in a B-tree is corrupted. I would expect the error message in a later release to report the file name and the page number or byte offset.

            marko Marko Mäkelä added a comment - Yes, a particular page in a B-tree is corrupted. I would expect the error message in a later release to report the file name and the page number or byte offset.
            Sangmin Sangmin Han added a comment - - edited

            Thank you for quick response.

            I have two important questions.

            1. Is there another way to check the above corrupt and prevent it in advance? (ex:innochecksum or mysqlcheck)

            2. Can the latest update to version 10.3 solve this issue?

            Sangmin Sangmin Han added a comment - - edited Thank you for quick response. I have two important questions. 1. Is there another way to check the above corrupt and prevent it in advance? (ex:innochecksum or mysqlcheck) 2. Can the latest update to version 10.3 solve this issue?
            Sangmin Sangmin Han added a comment -

            please answer about my question.

            Sangmin Sangmin Han added a comment - please answer about my question.
            Sangmin Sangmin Han added a comment -

            No updates for 3 months. Please let us know the current progress. And please answer a question I recently asked.

            Sangmin Sangmin Han added a comment - No updates for 3 months. Please let us know the current progress. And please answer a question I recently asked.

            I think that this particular type of crash should be prevented by MDEV-13542, which is currently being tested.

            This is a bug reporting system. MariaDB Corporation runs a separate system for providing technical support to customers.

            marko Marko Mäkelä added a comment - I think that this particular type of crash should be prevented by MDEV-13542 , which is currently being tested. This is a bug reporting system. MariaDB Corporation runs a separate system for providing technical support to customers.

            People

              marko Marko Mäkelä
              Sangmin Sangmin Han
              Votes:
              0 Vote for this issue
              Watchers:
              2 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.