• We value your experience with Plesk during 2024
    Plesk strives to perform even better in 2025. To help us improve further, please answer a few questions about your experience with Plesk Obsidian 2024.
    Please take this short survey:

    https://pt-research.typeform.com/to/AmZvSXkx
  • The Horde webmail has been deprecated. Its complete removal is scheduled for April 2025. For details and recommended actions, see the Feature and Deprecation Plan.
  • We’re working on enhancing the Monitoring feature in Plesk, and we could really use your expertise! If you’re open to sharing your experiences with server and website monitoring or providing feedback, we’d love to have a one-hour online meeting with you.

Bug: FTP is stopped during upgrade, but not restarted

Tozz

Regular Pleskian
All our machines running Plesk 11.5 with automatic updates enabled on Debian 6 are facing the issue that FTP is stopped during an automatic upgrade, but not started. Resulting in all servers having their FTP service shutdown.

Manually starting /etc/init.d/xinetd start resolves the issue. So its not actually FTP that is beeing shutdown, xinetd is shutdown and not restarted.

This happend about +/- 1 to 2 weeks ago and is now happening again.
 
Why do you think that it is related with microupdates installation? Do you have any proofs in log files?
 
Why do you think that it is related with microupdates installation? Do you have any proofs in log files?

I think that because it happens exactly when the autoinstaller is running, and there is proof of xinetd restarting. The other proof is that this happens on _all_ our machines running either Debian 6 or Debian 7. The logs below indicate that xinetd was started, but this seems false.

Please note:

Code:
root@plesk1:/tmp# cat /var/log/syslog | grep xinet
Aug 23 10:23:34 plesk1 xinetd[21549]: Exiting...
Aug 23 10:54:04 plesk1 xinetd[537]: Reading included configuration file: /etc/xinetd.d/chargen [file=/etc/xinetd.conf] [line=14]

This indicated that xinetd was stopped at 10:23:34 this morning. At 10:54 I manually restarted xinetd. Compare the 10:23 with the date code on the autoinstaller.log in /tmp:

Code:
root@plesk1:/tmp# ls -la /tmp/auto*
-rw------- 1 root root 5626543 Aug 23 10:24 /tmp/autoinstaller3.log

Both the matching timestamp, together with the fact that this happens on all our machines and that it seems to be recurring on every autoinstaller run I have no doubt this is caused by Parallels Autoinstaller/updater.

Code:
root@plesk1:/tmp# cat autoinstaller3.log  | grep xin
 Trying to add header to file /etc/xinetd.d/ftp_psa... file /etc/xinetd.d/ftp_psa already contains required header
 Trying to register service xinetd... done
 Trying to restart service xinetd... Stopping internet superserver: xinetd.
Starting internet superserver: xinetd.
 Trying to restart service xinetd... Starting internet superserver: xinetd failed!
 Trying to register service xinetd... done
 Trying to restart service xinetd... Stopping internet superserver: xinetd.
Starting internet superserver: xinetd.
Starting xinetd service... done
Starting xinetd service... done
Starting xinetd service... done
 Trying to add header to file /etc/xinetd.d/ftp_psa... file /etc/xinetd.d/ftp_psa already contains required header
 Trying to restart service xinetd... Starting internet superserver: xinetd.
Starting xinetd service... done
 Trying to add header to file /etc/xinetd.d/ftp_psa... file /etc/xinetd.d/ftp_psa already contains required header
 Trying to restart service xinetd... Starting internet superserver: xinetd.
 
We have to try reproduce it. If you need fast solution - contact Support Team. They are working in a 24x7
 
Thanks. I prefer the forums, since we're always running out of support incidents before the year ends. I am aware that bugreports should not be deducted from our number of remaining support incidents, but Parallels always says the bug was not known and must therefor be a user error. I got tired of it and started reporting issues here.
 
I would like to add that the autoinstaller auto seems to restart MySQL every time. Althought MySQL is restarted without issues, we monitor MySQL uptime (to detect corrupt tables that cause MySQL to crash) so this leads to an monitor alarm. I cannot find a good reason to restart MySQL during the installation of microupdates. It would be great if the useless MySQL restarts could be prevented
 
FYI: This bug resurfaced after today's MU26. All our FTP's are currently down after the auto updater started.
 
@thietkew: If you wo uld just take a look at my prior post, you will find the proof there.
 
I know it's been a while since last this thread was updated, but I have this same problem on a new plesk 12 debian server: Xinet stops after auto updates and does not start, the result being that ftp stops working.

For example, today:

The auto updates started ->
/tmp/autoinstaller3.log:
Parallels Installer 3.17.13 (built on 2014-12-17 12:38 svn rev. 337203) started at (timezone EET) Mon Feb 9 08:03:08 2015
...
...

and at the same time in daemon.log appears:

root@myserver ~ # grep xinetd /var/log/daemon.log
Feb 9 08:04:48 e-avenue xinetd[8774]: Exiting...

So, it is exiting, but not restarted. It looks quite like a bug. I attach the full relevant log of /tmp/autoinstaller3.log.
 

Attachments

  • plesk_installer_xinet_problem.txt
    154.1 KB · Views: 2
Unfortunately still present in
12.0.18 Update 49 [03-Jun-2015]

This has happened twice in the last few months after Plesk updates.
 
Back
Top