Uploaded image for project: 'MariaDB Server'
  1. MariaDB Server
  2. MDEV-4834

Multi source, can not create multiple domains on the same master

    Details

    • Type: Task
    • Status: Open
    • Priority: Minor
    • Resolution: Unresolved
    • Fix Version/s: 10.1
    • Component/s: None
    • Labels:

      Description

      Having a spider proxy to shard data on multiple schema into blackhole tables i now need some slave to read from the same master but using separate schema replicate-do-db. Every slave is replicating a couple of pre shard schema in asynchrone but in all case every slave can serve all the shards .

      change master 'dshard19' to master_host='192.168.0.30' , master_port=5013, master_user='skysql', master_password='skyvodka', master_log_file='mysql_sandbox5013-bin.000002', master_log_pos=6176;
      ERROR 1934 (HY000): Connection 'dshard19' conflicts with existing connection 'dshard01'

      What i don't get is during a prove of concept at a client we did not get this issue and have been able to build the same solution on the same release number only difference all instances was on dedicated servers .

      Can we introduce a parameter to enable multi source replication to a single master

        Gliffy Diagrams

          Attachments

            Activity

            Hide
            stephane@skysql.com VAROQUI Stephane added a comment -

            Using the workaround of <> hostname all pointing to the same master

            The error log is saturated with following errors we have 8 shards per physical servers * 2 messages * 4K inserts this is killing the disk We need a way to remove those messages. What we found is starting 9 replications domains and stoping the last one disable those messages and save us.

            30917 19:22:00 [Note] Master 'ccmstats_shard01': Slave I/O thread: Failed reading log event, reconnecting to retry, log 'mysql-bin.001701' at position 30233338
            130917 19:22:00 [Note] Master 'ccmstats_shard02': Slave: received end packet from server, apparent master shutdown:
            130917 19:22:00 [Note] Master 'ccmstats_shard02': Slave I/O thread: Failed reading log event, reconnecting to retry, log 'mysql-bin.001701' at position 30235938
            13

            Show
            stephane@skysql.com VAROQUI Stephane added a comment - Using the workaround of <> hostname all pointing to the same master The error log is saturated with following errors we have 8 shards per physical servers * 2 messages * 4K inserts this is killing the disk We need a way to remove those messages. What we found is starting 9 replications domains and stoping the last one disable those messages and save us. 30917 19:22:00 [Note] Master 'ccmstats_shard01': Slave I/O thread: Failed reading log event, reconnecting to retry, log 'mysql-bin.001701' at position 30233338 130917 19:22:00 [Note] Master 'ccmstats_shard02': Slave: received end packet from server, apparent master shutdown: 130917 19:22:00 [Note] Master 'ccmstats_shard02': Slave I/O thread: Failed reading log event, reconnecting to retry, log 'mysql-bin.001701' at position 30235938 13

              People

              • Assignee:
                Unassigned
                Reporter:
                stephane@skysql.com VAROQUI Stephane
              • Votes:
                2 Vote for this issue
                Watchers:
                3 Start watching this issue

                Dates

                • Created:
                  Updated: