• If you are still using CentOS 7.9, it's time to convert to Alma 8 with the free centos2alma tool by Plesk or Plesk Migrator. Please let us know your experiences or concerns in this thread:
    CentOS2Alma discussion

looking for log file (Plesk restart file)

W

wendy.mathis

Guest
I can no longer start Plesk on my server. In the details it tells me it failed but does not tell me why so I am not sure how to diagnose the issue.

I am assuming that the reason it failed it being written out to a log file but where?

I have just recently upgraded to plesk 8.2.1.
 
psa logs are in
/usr/local/psa/admin/logs

system logs are in
/var/log/messages

apache logs (not apache for plesk) are in
/var/log/httpd/error_log
 
None of the log files in that location have a modified date of today when i try to restart plesk. If it was writing to those log files, then wouldn't it have a modified date of the last errors? Meaning today when it failed....again.
 
Try and do this

Code:
/sbin/service psa stop
/sbin/service psa start
tail /usr/local/psa/admin/logs/httpsd_error_log
 
i tried that or rather i ran ./psa restart

this is what i got.

It is not writing to that error log. The entries in that file are from 2 days ago.

[root@localhost init.d]# ./psa restart PSA is down, performing full restart.
Starting xinetd service...
done
Starting named service...
done
Starting mysqld service...
done
Starting qmail service...
done
Starting courier-imap service...
done
Starting postgresql service...
not installed
Starting psa-spamassassin service...
done
Starting Plesk... failed
failed[root@l
[root@localhost init.d]# tail /usr/local/psa/admin/logs/httpsd_error_log
Downloading the file products.inf3: 100% done.Downloa
Downloading the file plesk.inf3: 100% done.Downloa
Downloading the file sitebuilder.inf3: 22%..100% done.Checkin
Checking for installed packages... ERROR: Autoinstaller busy
ERROR: Autoinstaller busy
[Tue Oct 9 15:58:58 2007] [warn] child process 12188 still did not exit, sending a SIGTERM
[Tue Oct 9 15:59:02 2007] [error] child process 12188 still did not exit, sending a SIGKILL
[Tue Oct 9 15:59:02 2007] [notice] caught SIGTERM, shutting down
 
Those log files were already posted and checked. This user says nothing is in them at all relating to psa
 
any more information on this topic? I have encountered the same issue, Plesk fails on

Code:
service psa startall

but writes nothing to any logs, drwebd also fails but does write to /var/log/messages
 
Hi.

service psa startall
will start all the services, like qmail, spamassassin, courier-imap, named etc..
You need to find out which one doesn't start.
Execute
/etc/init.d/psa start1
this will start only Plesk interface.
Does it start? is there any error messages?
If it doesn't start and you cannot see any errors, you can run it with:
sh -x /etc/init.d/psa start1
This will expand the errors and could give you a hint.
 
Thanks Monica

This at least gets me some errors and things to work with. i can start troubleshooting what is wrong.
 
IT WORKED!!!!

Thanks again Monica.

i actually tried to start the httpsd by itself and was able to see a real error message finally and once i got real error messages i could fix what was wrong.

Thanks again everyone for your help.
 
In the hopes it would help some one else, can you post what was wrong and what you did to fix?
 
I tried to start httpsdctl in the psa directory so that it evaluated the file for me.

Then i was getting errors becausue the httpsd.conf file had placeholders in it @admin_user@ and other such placeholders.

I replaced those values with the correct values for my system by looking at an old backup of the httpsd.conf file.

After updating that file and being able to successfully run the httpsdctl program, i then started psa normally and it was successful.

The key was being able to break down what the psa script was actually doing and then completely those steps manually to see where it failed.
 
Back
Top