After upgrade Plesk Panel 11.0.9 to 11.5 CentOS 6.4 hangs at reboot/shutdown

[SuppanP]

After performing an upgrade of Plesk Panel 11.0.9 to 11.5 on a CentOS 6.4 system running in a XEN VM the system crashes while rebooting/restarting (shutdown -r now). The VM can only be restarted via a forced shutdown of an administrator of my provider. The console log shows errors while unmounting disks and multiple tries to kill processes. Please see the attached console log for details (last pages - log shows the whole booting after upgrade and the failing shutdown process).

Up to now I have tried the upgrade four times (after reinstalling the previous snapshot) and the behavior is every time the same: After the upgrade the system starts one time regularly and the next shutdown ends with hanging up as shown in the log (seems to be a failure after the first reboot after the upgrade).

Before the upgrade the machines behavior was normal. The upgrade itself finished without any errors. No problems in the autoinstaller log.

What can i do? My provider (1&1 Germany) says that there is no problem with the virtual machine - it's a software problem which is out of their scope.

 

[IgorG]

This issue require serious and deep investigation directly on your server. Therefore I suggest you contact Parallels Support Team. If you have unlimited Plesk license - support will be free - http://www.parallels.com/products/plesk/get-support/free-support/

 

[hovik]

Likely a watchdog/psa-monit issue, see http://brablc.com/2013/09/21/plesk-watchdog-nginx-shutdown-problem/

The Plesk's self-monitoring system is so clever it restarts itself to ensure it's always running. Except when the server is shutting down, the brave die-hard self-monitoring system keeps restarting itself thus preventing your server from going down.

Welcome to the wonderful world of Plesk updates! I feel as part of Parallel's Quality Assurance Team again (if there is any). Throw up a shitty untested beta version and gather QA feedback from the users. Excellent policy!