Buy Percona ServicesBuy Now!

Help! Node won't join Cluster after Restart.

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Help! Node won't join Cluster after Restart.

    Hi Folks!

    I had my new XtraDB-Cluster (5.7.19, Debian 9) up and running for a while and today I decided to restart one of the three Nodes. That was apparently not a good Idea, because now it's not willing to rejoin the Cluster anymore. I really tried out what I could, but got the following Error, over and over again:

    "mysql[119540]: WSREP: Failed to recover position:"

    Here's my /etc/mysql/percona-xtradb-cluster.conf.d/mysqld.cnf which flawless works on the other 2 Nodes:

    wsrep_provider=/usr/lib/libgalera_smm.so
    wsrep_cluster_name=pxc-cluster
    wsrep_cluster_address=gcomm://xx.xx.107.133,xx.xx.107.134,xx.xx.107.135
    wsrep_node_name=myname
    wsrep_node_address=xx.xx.107.134
    wsrep_sst_method=xtrabackup-v2
    wsrep_sst_auth=sstuser:thepassword
    pxc_strict_mode=ENFORCING
    binlog_format=ROW
    default_storage_engine=InnoDB
    innodb_autoinc_lock_mode=2

    Also starting the Node, without making any changes to mysqld.cnf, and filling up the wsrep* variables manually (wsrep_cluster_address, global wsrep_provider, wsrep_node_address, wsrep_sst_auth) did not help.

    My Boss will be killing me, please help!

    With best regards,
    DBA Jay

  • #2
    Can you please share related mysql error log entries from joiner and donor node. Along with that, please share any innobackup* logs from joiner/donor data directory.

    Comment

    Working...
    X