Error 502 Bei Webmail und PMA

    Hallihallo,
    habe seit dem Update von 1.2.9 auf 1.2.17 das Problem, dass der Webmailer, sowie PMA Error 502 - Bad Gateway zeigen.


    Vorher gabs das Problem nicht, und das Problem besteht auf 3 Servern.
    Ich habe mir bereits andere Threads angeschaut, jedoch helfen Dinge wie das Erhöhen von den PHP_FCGI_CHILDREN in der /etc/imscp/nginx.data und [...]old.data nichts.


    Auch finde ich keine Logs die etwas aussagen. Ein Neustart des imscp_panel behebt das Problem für eine Weile, dann ist wieder alles wie vorher. (eine Weile = wenige Minuten bis Stunden)


    Die bisher nicht geupdateten 1.2.9er laufen super.


    Was für infos benötigt ihr noch?


    Grüße
    nomix

    Systeminfos bitte.. Hast du automatisch zu Ubuntu 16.04 geupgraded, entsprechend deine Packages...?

    Ohja stimmt verbimmelt.


    Debian 8 - Kernel 3.16


    Hier die vollständige Paketliste:

    Display Spoiler


    Noch weitere Infos? Die Ngnix logs sind leer.

    Update entsprechend mal auf den 1.2.x branch.

    @nomix


    You can update using https://github.com/i-MSCP/imscp/archive/1.2.x.zip
    The process is identical. But please first, read the changelog and the errata file.

    badge.php?id=1239063037&bid=2518&key=1747635596&format=png&z=547451206

    SSH key for online support

    Without further info, we cannot help you. Give us access to your server. Then we can investigate.

    badge.php?id=1239063037&bid=2518&key=1747635596&format=png&z=547451206

    SSH key for online support

    @nomix


    I investigated your server. Here come the reasons:


    First, I've checked the status of the imscp_panel service which is responsible to spawn the PHP processes for the i-MSCP frontEnd:


    Code
    1. ● imscp_panel.service - i-MSCP frontEnd PHP processesLoaded: loaded (/etc/systemd/system/imscp_panel.service; enabled)Active: inactive (dead) since jeu. 2016-05-19 03:16:09 CEST; 4h 8min agoProcess: 21981 ExecStart=/var/www/fcgi/master/php-fcgi-starter (code=exited, status=0/SUCCESS)Process: 21978 ExecStartPre=/bin/rm -f /var/run/imscp_panel.sock (code=exited, status=0/SUCCESS)Main PID: 21983 (code=exited, status=0/SUCCESS)mai 19 01:39:27 herodes php-fcgi-starter[21981]: spawn-fcgi: child spawned successfully: PID: 21983


    As you can see here, the service is dead since jeu. 2016-05-19 03:16:09 CEST; 4h 8min ago. Thus, with this information in the hands, I've looked at your previous syslog /var/log/syslog.1 file (which has been rotated this morning) and, I've found the following logs:

    Code
    1. ...May 19 03:16:01 herodes CRON[28431]: (root) CMD (/root/secusys/system >/dev/null 2>&1)May 19 03:16:03 herodes apache2[28441]: Stopping web server: apache2.May 19 03:16:04 herodes apache2[28466]: Starting web server: apache2.May 19 03:16:05 herodes proftpd[20989]: 8x.1xx.247.xx - ProFTPD killed (signal 15)May 19 03:16:05 herodes proftpd[20989]: 8x.1xx.247.xx - ProFTPD 1.3.5 standalone mode SHUTDOWNMay 19 03:16:05 herodes proftpd[28557]: Stopping ftp server: proftpd.May 19 03:16:05 herodes proftpd[28574]: 8x.1xx.247.xx - ProFTPD 1.3.5 (stable) (built Tue May 19 2015 11:09:58 UTC) standalone mode STARTUPMay 19 03:16:05 herodes proftpd[28567]: Starting ftp server: proftpd.May 19 03:16:07 herodes mysql[28585]: Stopping MySQL database server: mysqld.May 19 03:16:09 herodes mysql[28680]: Starting MySQL database server: mysqld ..May 19 03:16:09 herodes mysql[28680]: Checking for tables which need an upgrade, are corrupt or wereMay 19 03:16:09 herodes mysql[28680]: not closed cleanly..May 19 03:16:09 herodes /etc/mysql/debian-start[29162]: Upgrading MySQL tables if necessary.May 19 03:16:09 herodes /etc/mysql/debian-start[29166]: /usr/bin/mysql_upgrade: the '--basedir' option is always ignoredMay 19 03:16:09 herodes /etc/mysql/debian-start[29166]: Looking for 'mysql' as: /usr/bin/mysqlMay 19 03:16:09 herodes /etc/mysql/debian-start[29166]: Looking for 'mysqlcheck' as: /usr/bin/mysqlcheckMay 19 03:16:09 herodes /etc/mysql/debian-start[29166]: This installation of MySQL is already upgraded to 5.5.49, use --force if you still need to run mysql_upgradeMay 19 03:16:09 herodes /etc/mysql/debian-start[29180]: Checking for insecure root accounts.May 19 03:16:09 herodes /etc/mysql/debian-start[29185]: Triggering myisam-recover for all MyISAM tables...


    So here, we can see that you run the /root/secusys/system script through a cron job. This is confirmed by;


    Code
    1. root@herodes:/var/log# crontab -l# Edit this file to introduce tasks to be run by cron.## Each task to run has to be defined through a single line# indicating with different fields when the task will be run# and what command to run for the task## To define the time you can provide concrete values for# minute (m), hour (h), day of month (dom), month (mon),# and day of week (dow) or use '*' in these fields (for 'any').## Notice that tasks will be started based on the cron's system# daemon's notion of time and timezones.## Output of the crontab jobs (including errors) is sent through# email to the user the crontab file belongs to (unless redirected).## For example, you can run a backup of all your user accounts# at 5 a.m every week with:# 0 5 * * 1 tar -zcf /var/backups/home.tgz /home/## For more information see the manual pages of crontab(5) and cron(8)## m h dom mon dow command@reboot /etc/init.d/xinetd restart &> /dev/null16 3 * * * /root/secusys/system >/dev/null 2>&10 */1 * * * /root/secusys/backup.sh >/dev/null 2>&1


    The problem is in the /root/secusys/system script itself which has the following content:



    As you can see here, you're killing all php5-cgi processes. That is really bad. Doing this also kills the PHP processes that are spawned by the imscp_panel service. This explain why you get those 502 errors.



    I would really recommend you to review that script which is anyway really bad. You must never kill PHP processes in that way. That script should finish into the toilets ;)


    In order, I would recommend to simply remove that script which is totally useless:

    • It is not needed to restart service (apache2, proftpd, mysql).
    • It is not needed to perform mysql optimization as you do
    • It is not needed to kill PHP processes as you do (even worse, this can lead to serious side-effects)

    I would also recommend you to switch to PHP FPM (you currently use fcgid).

    badge.php?id=1239063037&bid=2518&key=1747635596&format=png&z=547451206

    SSH key for online support