0

I am in the process of trying to set up a MariaDB galera cluster. The first 2 nodes went fine, have them started, connected.

On the third node, it won't start/connect. 

# service mysql start
Starting mysql (via systemctl):  Job for mariadb.service failed because a timeout was exceeded. See "systemctl status mariadb.service" and "journalctl -xe" for details.
                                                           [FAILED]

journal -xe output:

Jan 19 09:16:07 host3.domain.com systemd[1]: mariadb.service start operation timed out. Terminating.
-- Subject: Unit session-c9591.scope has finished start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- The start-up result is done.

-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit session-c9591.scope has begun starting up.
Jan 19 09:17:01 host3.domain.com CROND[1018]: (root) CMD (/usr/local/rtm/bin/rtm 8 > /dev/null 2> /dev/null)
Jan 19 09:17:38 host3.domain.com systemd[1]: mariadb.service stop-final-sigterm timed out. Skipping SIGKILL. Entering failed mode.
Jan 19 09:17:38 host3.domain.com systemd[1]: Failed to start MariaDB 10.1.30 database server.
-- Subject: Unit mariadb.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit mariadb.service has failed.
-- 
-- The result is failed.
Jan 19 09:17:38 host3.domain.com systemd[1]: Unit mariadb.service entered failed state.
Jan 19 09:17:38 host3.domain.com systemd[1]: mariadb.service failed.
Jan 19 09:17:38 host3.domain.com polkitd[383]: Unregistered Authentication Agent for unix-process:25848:56441890 (system bus name :1.19233, object path /org/freedesktop/PolicyKit1/Authentic
Jan 19 09:17:51 host3.domain.com mysqld[25932]: 2018-01-19  9:17:51 114327532205824 [Note] WSREP: (15573658, 'tcp://0.0.0.0:4567') connection to peer 00000000 with addr  timed out, no messa
Jan 19 09:18:01 host3.domain.com systemd[1]: Started Session c9592 of user root.
-- Subject: Unit session-c9592.scope has finished start-up
-- Defined-By: systemd

I am not clear on why it is timing out why connecting to itself? Is there another log file I should generate that would offer more clues? I confirmed in syslog that the error 

[Note] WSREP: (15573658, 'tcp://0.0.0.0:4567') connection to peer 00000000 with addr  timed out, no messa

the missing addr is internal private IP.

Marc
  • 33
  • 11

1 Answers1

1

For me the fix was simply to tear down the entire cluster and start it back up.

on each node:

service mysql stop

On the most advanced node

# galera_new_cluster

On each subsequent node

service mysql start

Verify with:

MariaDB [(none)]> show global status like "%wsrep_cluster_size%";

+--------------------+-------+
| Variable_name      | Value |
+--------------------+-------+
| wsrep_cluster_size | 3     |
+--------------------+-------+
1 row in set (0.00 sec)
Marc
  • 33
  • 11