Uploaded image for project: 'MariaDB Server'
  1. MariaDB Server
  2. MDEV-35290

galera_new_cluster mariadb@<instnace_name> failed when wsrep_cluster_address already set some nodes address

Details

    • Bug
    • Status: Open (View Workflow)
    • Major
    • Resolution: Unresolved
    • 10.6
    • 10.6
    • Galera
    • Linux using systemd

    Description

      I use "galera_new_cluster mariadb@node1" to bootstap a new cluster.

      If set "wsrep_cluster_address="gcomm://" in config file, it is ok to startup a new cluster,
      If set "wsrep_cluster_address="gcomm://10.10.10.1:4567,10.10.10.2:4567,10.10.10.3:4567" in config file, it will fail startup a new cluster.

      I find a way to fix this problem:
      edit "/usr/lib/systemd/system/mariadb@.service" file and append $_WSREP_NEW_CLUSTER in the end of "ExecStart="
      then mariadb@.service can catch the environment variable _WSREP_NEW_CLUSTER="--wsrep-new-cluster" (setting in "galera_new_cluster" script)

      After a "systemctl daemon-reload", then "galera_new_cluster mariadb@node1" will work fine to startup a new cluster.

      Attachments

        Activity

          No workflow transitions have been executed yet.

          People

            janlindstrom Jan Lindström
            garfield JerShiuh Lin
            Votes:
            0 Vote for this issue
            Watchers:
            2 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.