Uploaded image for project: 'MariaDB Server'
  1. MariaDB Server
  2. MDEV-36237

mariabackup fails with 'try increasing innodb_log_file_size'

Details

    • Bug
    • Status: Open (View Workflow)
    • Major
    • Resolution: Unresolved
    • 11.4.4
    • 11.4
    • Backup
    • None

    Description

      The backups would fail with this error despite multiple increases to the innodb_log_file_size.

      [00] 2025-03-03 18:10:25 Was only able to copy log from 63677889668553 to 63680238996661, not 63682682984728; try increasing innodb_log_file_size
      		 
      mariabackup: Stopping log copying thread.[00] 2025-03-03 18:10:25 Retrying read of log at LSN=63680238996661

      Eventually, after increasing the innodb_log_file_size to 30G, the backups began working.

      This is the backup command used:

      mariabackup --backup --kill-long-queries-timeout=300 --kill-long-query-type=SELECT --open-files-limit=100000 --user=backup --password=x xxxxxxxxxxxxxxxxxx --socket=/run/mysqld/mysqld.sock --parallel=4 --extra-lsndir=/tmp --stream=xbstream --slave-info --ftwrl-wait-timeout=30 --ftwrl-wait-threshold=999999 --target-dir=/backup//2025-03-03/mariabackup

      Attachments

        Issue Links

          relates to

          Bug - A problem which impairs or prevents the functions of the product. MDEV-35442 mariabackup 10.11 failed

          • Major - Major loss of function.
          • Closed

          Activity

            Ascending order - Click to sort in descending order
            bwillits Bill Willits created issue -
            bwillits Bill Willits made changes -
            Field Original Value New Value
            Description The backups would fail with this error despite multiple increases to the innodb_log_file_size.
            {code:java}
            [00] 2025-03-03 18:10:25 Was only able to copy log from 63677889668553 to 63680238996661, not 63682682984728; try increasing innodb_log_file_size
            mariabackup: Stopping log copying thread.[00] 2025-03-03 18:10:25 Retrying read of log at LSN=63680238996661
            {code}
            Eventually, after increasing the innodb_log_file_size to 30G, the backups began working.             		The backups would fail with this error despite multiple increases to the innodb_log_file_size.
            {code:java}
            [00] 2025-03-03 18:10:25 Was only able to copy log from 63677889668553 to 63680238996661, not 63682682984728; try increasing innodb_log_file_size
            mariabackup: Stopping log copying thread.[00] 2025-03-03 18:10:25 Retrying read of log at LSN=63680238996661
            {code}
            Eventually, after increasing the innodb_log_file_size to 30G, the backups began working.

            This is the backup command used:
            {noformat}
            mariabackup --backup --kill-long-queries-timeout=300 --kill-long-query-type=SELECT --open-files-limit=100000 --user=backup --password=x xxxxxxxxxxxxxxxxxx --socket=/run/mysqld/mysqld.sock --parallel=4 --extra-lsndir=/tmp --stream=xbstream --slave-info --ftwrl-wait-timeout=30 --ftwrl-wait-threshold=999999 --target-dir=/backup//2025-03-03/mariabackup
            {noformat}

            bwillits Bill Willits made changes -
            Description The backups would fail with this error despite multiple increases to the innodb_log_file_size.
            {code:java}
            [00] 2025-03-03 18:10:25 Was only able to copy log from 63677889668553 to 63680238996661, not 63682682984728; try increasing innodb_log_file_size
            mariabackup: Stopping log copying thread.[00] 2025-03-03 18:10:25 Retrying read of log at LSN=63680238996661
            {code}
            Eventually, after increasing the innodb_log_file_size to 30G, the backups began working.

            This is the backup command used:
            {noformat}
            mariabackup --backup --kill-long-queries-timeout=300 --kill-long-query-type=SELECT --open-files-limit=100000 --user=backup --password=x xxxxxxxxxxxxxxxxxx --socket=/run/mysqld/mysqld.sock --parallel=4 --extra-lsndir=/tmp --stream=xbstream --slave-info --ftwrl-wait-timeout=30 --ftwrl-wait-threshold=999999 --target-dir=/backup//2025-03-03/mariabackup
            {noformat}

            		The backups would fail with this error despite multiple increases to the innodb_log_file_size.
            {noformat}
            [00] 2025-03-03 18:10:25 Was only able to copy log from 63677889668553 to 63680238996661, not 63682682984728; try increasing innodb_log_file_size
            mariabackup: Stopping log copying thread.[00] 2025-03-03 18:10:25 Retrying read of log at LSN=63680238996661
            {noformat}
            Eventually, after increasing the innodb_log_file_size to 30G, the backups began working.

            This is the backup command used:
            {noformat}
            mariabackup --backup --kill-long-queries-timeout=300 --kill-long-query-type=SELECT --open-files-limit=100000 --user=backup --password=x xxxxxxxxxxxxxxxxxx --socket=/run/mysqld/mysqld.sock --parallel=4 --extra-lsndir=/tmp --stream=xbstream --slave-info --ftwrl-wait-timeout=30 --ftwrl-wait-threshold=999999 --target-dir=/backup//2025-03-03/mariabackup
            {noformat}

            serg Sergei Golubchik added a comment - - edited

            marko, a duplicate of MDEV-34850?

            serg Sergei Golubchik added a comment - - edited marko , a duplicate of MDEV-34850 ?
            marko Marko Mäkelä added a comment -

            I suspect that this could basically be a design issue that would ultimately be fixed by implementing some form of backup in the server (MDEV-14992). Starting with MDEV-27774 the disparity between backup and server grew larger, because the server may produce log from multiple concurrent threads, but backup can only read and parse it in a single thread.

            In some internal discussion about this ticket, also incremental backup was discussed, but I did not dive into details about that. In any case, MDEV-30000 might provide some help for this problem.

            marko Marko Mäkelä added a comment - I suspect that this could basically be a design issue that would ultimately be fixed by implementing some form of backup in the server ( MDEV-14992 ). Starting with MDEV-27774 the disparity between backup and server grew larger, because the server may produce log from multiple concurrent threads, but backup can only read and parse it in a single thread. In some internal discussion about this ticket, also incremental backup was discussed, but I did not dive into details about that. In any case, MDEV-30000 might provide some help for this problem.
            serg Sergei Golubchik made changes -
            Assignee Marko Mäkelä [ marko ]
            serg Sergei Golubchik made changes -
            Fix Version/s 11.4 [ 29301 ]
            Richard Richard Stracke made changes -

            People

              marko Marko Mäkelä
              bwillits Bill Willits
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

                Created:
                Updated:

                Git Integration

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