Uploaded image for project: 'MariaDB MaxScale'
  1. MariaDB MaxScale
  2. MXS-2228

GTID values not updated if only slaves are present

    XMLWordPrintable

Details

    • Bug
    • Status: Closed (View Workflow)
    • Major
    • Resolution: Not a Bug
    • 2.2.18
    • 2.3.0
    • mariadbmon
    • None
    • Reproduced with MariaDB 10.1.36 on docker

    Description

      When MaxScale is configured to monitor the last two slaves (S2 and S3) of a four node chained cluster:

      M <- S1 <- S2 <- S3

      The GTID values in the maxctrl list servers are not shown when the slave servers have read-only on. If read-only is not enabled, S2 is labeled as the master and the GTIDs are updated.

      This is caused by the requirement to know the master's domain ID before the GTID update is done (line 1814 in mariadbmon.cc). By unconditionally updating the GTID positions of the servers, the GTIDs are updated correctly even when only slaves are being monitored.

      Reproduced with this configuration

      [server3]
      type=server
      address=127.0.0.1
      port=3002
      protocol=MariaDBBackend
       
      [server4]
      type=server
      address=127.0.0.1
      port=3003
      protocol=MariaDBBackend
       
      [MariaDB-Monitor]
      type=monitor
      module=mariadbmon
      servers=server3,server4
      user=maxuser
      password=maxpwd
      monitor_interval=2000
       
      [CLI]
      type=service
      router=cli
       
      [CLI-Listener]
      type=listener
      service=CLI
      protocol=maxscaled
      socket=default
      

      Attachments

        Activity

          People

            esa.korhonen Esa Korhonen
            markus makela markus makela
            Votes:
            1 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.