Details
Minor Description
On startup the image throws the following warnings, each of which seems to contribute to the container taking longer to finish start-up tasks, making full start-up take over a minute.
Least impactful is the first warning-
2021-11-17 20:47:02 1 [Warning] Failed to load slave replication state from table mysql.gtid_slave_pos: 1017: Can't find file: './mysql/' (errno: 2 "No such file or directory")
|
This one is probably fine as-is, but if it can be addressed, would be nice.
The bigger issues are these three-
2021-11-17 20:47:07 0 [Note] mariadbd: ready for connections.
|
Version: '10.6.5-MariaDB-1:10.6.5+maria~focal-log' socket: '/run/mysqld/mysqld.sock' port: 0 mariadb.org binary distribution
|
2021-11-17 20:47:07+00:00 [Note] [Entrypoint]: Temporary server started.
|
Warning: Unable to load '/usr/share/zoneinfo/leap-seconds.list' as time zone. Skipping it.
|
Warning: Unable to load '/usr/share/zoneinfo/leapseconds' as time zone. Skipping it.
|
Warning: Unable to load '/usr/share/zoneinfo/tzdata.zi' as time zone. Skipping it.
|
2021-11-17 20:47:57 6 [Warning] 'proxies_priv' entry '@% root@3ca6dd5b620d' ignored in --skip-name-resolve mode.
|
2021-11-17 20:47:57+00:00 [Note] [Entrypoint]: Creating database test
|
2021-11-17 20:47:57+00:00 [Note] [Entrypoint]: Creating user tester
|
2021-11-17 20:47:57+00:00 [Note] [Entrypoint]: Giving user tester access to schema test
|
2021-11-17 20:47:57+00:00 [Note] [Entrypoint]: Stopping temporary server
|
2021-11-17 20:47:57 0 [Note] mariadbd (initiated by: root[root] @ localhost []): Normal shutdown
|
2021-11-17 20:47:57 0 [Note] InnoDB: FTS optimize thread exiting.
|
Key here are the time stamps- it is 20:47:07 before the Warning: Unable to load... entries occur and then 20:47:57 after. It seems like the system waits for these to time out before proceeding.
The amount of time here doesn't look significant, but this was a faster than typical run for this- it can take longer. It also feels inordinately long compared to when the temporary server starts up (and also quite long compared to Nginx and PHP-FPM container start-up times, making MariaDB the bottleneck for launching new EMP stacks). If we can resolve or work-around this issue, the MariaDB Docker container will feel much more responsive and yield a much improved user experience.
Attachments
Issue Links
- causes
-
MDEV-28103 container /docker-entrypoint-initdb.d sql on master doesn't replicate
-
- Closed
-
- relates to
-
-
- Closed
-
-
MDEV-27167 system variable deferred-networking to resolve early initialization
-
- Closed
-
-
-
- Closed
-