Uploaded image for project: 'MariaDB Server'
  1. MariaDB Server
  2. MDEV-33258

Lacks persistant config options specificly for default channel under multi source replications

Details

    • Bug
    • Status: Open (View Workflow)
    • Major
    • Resolution: Unresolved
    • 11.1.2
    • 11.1(EOL)
    • None

    Description

      Lacks persistent config options specifically for default channel under multi source replications
      I assume it affects all versions however i only tested in 11.1.2

      in mariadb non specified variable channels in the config file for replications automatically binds to all other channels.

      There is no way to specifically set a variable to only be used on the default channel

      The only workaround i have found is to set the same variable as a dummy variable to over set the default variable

      I also can not set a null value in the config file to reset it

      example:

      [mariadbd-11.1]
      // Default channel
      replicate-do_table=test.test_log  ##  There is no way to make this variable only for default
      replicate_do_table=test.admin_log
      replicate_do_table=test.account_managers
      replicate_do_table=test.banks
       
      // company channel
       
      temp.replicate_ignore_db=test
      temp.replicate_wild_do_table="k_company_%.%"
      temp.replicate_do_table='k_company_testers_4_life.%'   #####  i have to specify something here to remove the above variables
       
      temp.replicate_do_table=''   ##  setting this in cli  works with the desired effects  ,  however  this crashes mariadb  if there is no input
      

      Attachments

        Activity

          Hi Ford!

          Our documentation on the subject mentions this:

          If one doesn't use any connection name prefix for a replicate.. variable, then the value will be used as the default value for all connections that don't have a value set for this variable.

          with the intention to help aid in configuring multiple channels with shared variable values. Generally, with multi-source replication, it'd be best if your connections all were named, and the default connection was used just for the "global configuration" baseline. Then each named connection in your my.cnf would override those global defaults.

          I'll also look into the null report, that seems broken

          bnestere Brandon Nesterenko added a comment - Hi Ford ! Our documentation on the subject mentions this: If one doesn't use any connection name prefix for a replicate.. variable, then the value will be used as the default value for all connections that don't have a value set for this variable. with the intention to help aid in configuring multiple channels with shared variable values. Generally, with multi-source replication, it'd be best if your connections all were named, and the default connection was used just for the "global configuration" baseline. Then each named connection in your my.cnf would override those global defaults. I'll also look into the null report, that seems broken

          People

            bnestere Brandon Nesterenko
            Ford Allen
            Votes:
            0 Vote for this issue
            Watchers:
            3 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.