Went through howtoforge search, read threads but no help for my single server setup that only does mail. Managed to almost cut down all the mysql warnings with the help 1, 2 Now the last days i do always see this (below) last remaining warning repeat at the same time UTC 22:25 and then around UTC22:53 or UTC22:55 and it is always 9 warnings both times. Not seeing any thing else happening in syslog prior to the warnings, any ideas where to look next? Code: May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22037 [Warning] Aborted connection 22037 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22034 [Warning] Aborted connection 22034 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22031 [Warning] Aborted connection 22031 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22030 [Warning] Aborted connection 22030 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22032 [Warning] Aborted connection 22032 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22038 [Warning] Aborted connection 22038 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22040 [Warning] Aborted connection 22040 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22039 [Warning] Aborted connection 22039 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) May 20 22:25:45 mail2 mariadbd[422]: 2023-05-20 22:25:45 22033 [Warning] Aborted connection 22033 to db: 'dbispconfig' user: 'ispconfig' host: 'localhost' (Got an error reading communication packets) changes added to 50-server.cnf Code: :~# grep -vE '^#|^ *$' /etc/mysql/mariadb.conf.d/50-server.cnf [server] [mysqld] pid-file = /run/mysqld/mysqld.pid basedir = /usr bind-address = 127.0.0.1 max_allowed_packet = 256M max_connections = 400 interactive_timeout = 6000 wait_timeout = 6000 net_read_timeout = 6000 net_write_timeout = 6000 expire_logs_days = 10 character-set-server = utf8mb4 collation-server = utf8mb4_general_ci sql-mode = "NO_ENGINE_SUBSTITUTION" [embedded] [mariadb] [mariadb-10.6]
This is nothing to worry about if the system works and behaves like expected. The error can occure when, a client, in this case ispconfig itself, terminates the connection improperly; The client was in sleep state longer than the wait_timout is configured for the DB; Or when the script expcted or exceeded the DBs max_allowed_paket configuration. I wouldn't worry to much if this is in the logs from time to time and everything is working.
The annoying part is that it shows up daily at the same time even made sure yesterday to reboot the server to see if the UTC time of warnings would change but the time they occur remains the same UTC 22:25 and then 22:50. Made me wonder and at the same time thought might be interesting for devs to look into it. Thanks for the reply.
I've just checked on some of my servers but there is nothing that shows any regularity. On some the message occurred at some point but it's rare. Do you have anything setup or running beside a normal ISPConfig Installation? May @till has some insight, and knows if there is something happening at special times that could trigger this.
Look at the ISPConfig cron plugins in /usr/local/ispconfig/server/lib/classes/cron.d/.Ecah file has a line like this: Code: protected $_schedule = '*/20 * * * *'; Which shows when it gets executed. The syntax is the same as normal Linux cronjobs use. maybe you find a plugin there that matches your time pattern to narrow things down.
Second Night now without seeing the Warning the only change i did was some config changes to an lan ssmtp server conf since after the migration/upgrade (rspamd now) the non-root cron job emails were flagged as spam. For more context: That lan server had a user account running a cron job which had an error and was informing me via email around that time. There was actually not much miss configured in ssmtp just took out some LOGIN line the AUTH lines were enough for it to work. I would actually have to reproduce the error (forgot to backup previous ssmtp conf) before i could say that miss configured ssmtp could cause the Warning but i am looking for another solution and probably will use mutt for the non-root users. I know it's off topic any hints though are welcome
