Have website I do not longer need. website has shell user. would make new shell user with same name on other website. Got error shell user with same name exists. So I decided to delete the shell user and after that deleting the web site. Now 15 Min after deleting the shell user it st still not finished, the red number still blinking.Job queue lists still the job. No errors in system log, server up accessable via ssh , nearly not load, other websites on that server running. Multiserver system ISPconfig runs on an other server. need hint, thanks Rainer
Could do other things on that server before i.e. creating website. delete still runing about 45 minutes now
Following errors PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 80 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 I deletet an other shell user on same server just before the action creating the error. As I wrote other actions before on the server worked fine. mysql ist running on all servers , reboot of server with the ISPconfig master db and reboot of the affected server no change. the affected server should be remove ist 1 last website well bemoved, but I have to get rid of the blocking action before. Other question, if a move a website, that means backup files an database, delete it, recreate it an restore everything, can I use the old shell user (jailed) or have I to delete an recreate it.
According to the error, ISPConfig is not able to connect to its mysql database. Did you change anything in the mysql configuration on that server or on the master before the issue started, e.g. did you disable name lookups or something similar?
No, I can not remember. Specially as I could delete an other shell user some minutes before. Can you give me a hint how to get nearer to the problem? Places (conf files etc,) I can look, test I can do. Thanks Rainer
Please use the debug mode: https://www.faqforge.com/linux/debugging-ispconfig-3-server-actions-in-case-of-a-failure/ enable it on the master for the affected server and then run server.sh and post the output that you get on the shell.
As I understand I stopp server.sh cronjob on the master and run server.sh on the master. That shows me not really much root@admin:/usr/local/ispconfig/server# ./server.sh finished. root@admin:/usr/local/ispconfig/server# Checked mysql, ran systemctl daemon-reload, checked again root@admin:/usr/local/ispconfig/server# service mysql status Warning: mariadb.service changed on disk. Run 'systemctl daemon-reload' to reload units. ● mariadb.service - MariaDB 10.1.38 database server Loaded: loaded (/lib/systemd/system/mariadb.service; enabled; vendor preset: enabled) Active: active (running) since Mon 2019-05-13 10:12:20 CEST; 4 days ago Docs: man:mysqld(8) https://mariadb.com/kb/en/library/systemd/ Main PID: 565 (mysqld) Status: "Taking your SQL requests now..." CGroup: /system.slice/mariadb.service └─565 /usr/sbin/mysqld Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable. How can I get rid off the blocking job? Thanks for help Rainer
It is a Multiserver system the Master Server who doe primary DNS too Two Web Server, one of them ist failing to remove the shell User, and Mail Server and so on on the failing Server I get, as posted above root@http2:~# /usr/local/ispconfig/server/server.sh 17.05.2019-16:12 - WARNING - There is already a lockfile set, but no process running with this pid (1658). Continuing. PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 80 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 who/where do I finde the lockfile for the pid 1658, find / -name "*1658*" as root no success Need at least hint to find reason for the warning. As I wrote too the task I did just some minutes before the failure was successfully delete an other shell user on the same web server Thanks Rainer
It is a Multiserver system the Master Server who doe primary DNS too Two Web Server, one of them ist failing to remove the shell User, and Mail Server and so on on the failing Server I get, as posted above root@http2:~# /usr/local/ispconfig/server/server.sh 17.05.2019-16:12 - WARNING - There is already a lockfile set, but no process running with this pid (1658). Continuing. PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 80 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 PHP Warning: mysqli_real_connect(): (HY000/2002): Connection refused in /usr/local/ispconfig/server/lib/classes/db_mysql.inc.php on line 83 who/where do I finde the lockfile for the pid 1658, find / -name "*1658*" as root no success Need at least hint to find reason for the warning. As I wrote too the task I did just some minutes before the failure was successfully delete an other shell user on the same web server Thanks Rainer
you don't have to as the script told you that it ignored the stale lock file. Your problem is that the slave server is not able to connect to mysql anymore and therefore, changes cannot be processed. This is just a timely coincidence and not related to the problem. Do you remember what you changed in mysql configuration on master or slave server or the firewall on master which prevents mysql logins now?
No changes on firewall copied the master from debian 8 to debian 9 with copy tool, after building the new master useing you turorials an manuals . Think, quite shure I deleted an other webserver succsessfully after the copy. I do not have big problem to delete der problem webserver as ist should be deleted anyway, but have to move a last existing server before. But the sticky task blocks everything, so again, can I delete the sticky task. Who can I test the connection from the slave to the master, If I know the excat reason I think I can fix it. Thanks
There is no sticky task that blocks anything and also other nodes are not affected when one node fails to connect to the master. Your problem is that the slave server cannot connect to master mysql database anymore and therefore, no tasks can be fetched by that slave and they stay in the queue. Test the MySQL connection from slave to master by using the dbmaster login details from /usr/local/ispconfig/server7lib/config.inc.php on the slave. You can test the connection with the mysql command on the shell. if this connection fails, then the slave cannot process changes and they get stored in the queue until the slave is able to connect again. you can also try to run an ispconfig update on the slave with ispconfig_update.sh and choose to reconfigure permissions in master database during update.
OK found problem, on of the slave servers can connect master, manual remote mysql root login to master not, possible, but local login works. found my.cnf is gone and replaced with a link to a nearly empty alternate my.conf. Will recreate the my.cnf. I think I can copy it from an other server Thanks Rainer
Correction, my.cnf is OK, just the comment at the bind address in /etc/mysql/mariadb.conf.d/50-server.cnf was lost. The bigger problem is the user table from mysql db is nearly empty, just root, ispconfig and phpadmin user still exist. While the ispconfig db seams to be OK as I can see all settings in the ispconfig web site at port 8080. I'll have no problem to recreate the root login entries as discribed in the 3.1 manual begining page 30 and following. What about ispsrvX users, have I create them manually with passwords from /usr/local/ispconfig/server/lib/config.inc.php or will ispconfig_update.sh on slave servers recreate them. Need some hints what to do, will not destroy more than what is still destroyed. Also the phpadmin root login fails, the recreation of the users works on command line too, but who to reenable php admin? Thanks Rainer
The ispcsrv* users exist only on the master, the ispconfig_update.sh script should recreate them when you run it on the slave, choose yes when the updater asks if it shall reconfigure permissions on master.
Thank you very much for help and hints, everything seams to run fine again. Still wondering how, on master server, most entries in the mysql.user table are gone lost. So all root@ip entries, [email protected] entries, ispcsrvX entries, and the root@localhost entry (phpmyadmin could not log anymore with root@localhost), while the ispconfig database was totally ok. After recreating the root@ipaddress and [email protected] entries and the root@localhost as running the ispconfig_update.sh on every slave with reconfigure permission the jobqueue was empty, the shell user deleted. Rainer
