CentOS 6 64bit + Plesk 11.5 update to 12.0.18

[GharibK]

i have a dedicated server and am running CentOS 6 64bit + Plesk 11.5 and updated last night my server to plesk 12.0.18

after update i got an notification:

New configuration files for the Apache web server were not created due to the errors in configuration templates: Can not restart web server: Service /etc/init.d/httpd failed to start Service /etc/init.d/httpd failed to start . Detailed error descriptions were sent to you by email. Please resolve the issues and click here to generate broken configuration files once again or here to generate all configuration files. See the details in Configuration Troubleshooter

i tried both options but it didnt work.
option 1 got reaction:

Error: Web server reconfiguration failed: New configuration was not built.

option 2 got reaction:

Unable to configure the web server: Execution failed. Command: httpdmng Arguments: Array ( [0] => --reconfigure-all ) Details: Execution failed. Command: httpdmng Arguments: Array ( [0] => --reconfigure-server [1] => -no-restart ) Details: [2014-09-12 12:57:20] ERR [util_exec] proc_close() failed [2014-09-12 12:57:24] ERR [util_exec] proc_close() failed [2014-09-12 12:57:25] ERR [panel] Apache config (14105194370.65768500) generation failed: Template_Exception: nginx: [emerg] PEM_read_bio_X509_AUX("/usr/local/psa/var/certificates/certFATV5bZ") failed (SSL: error:0906D064EM routinesEM_read_bio:bad base64 decode) nginx: configuration file /etc/nginx/nginx.conf test failed file: /usr/local/psa/admin/plib/Template/Writer/Webserver/Abstract.php line: 75 code: 0 nginx: [emerg] PEM_read_bio_X509_AUX("/usr/local/psa/var/certificates/certFATV5bZ") failed (SSL: error:0906D064EM routinesEM_read_bio:bad base64 decode) nginx: configuration file /etc/nginx/nginx.conf test failed

i tried to login with ssh and reinstall the webserver but that didnt work either.

i was reading on the forum that someone fixed this problem by recreating a new ssl certificate but that didnt work with me.

anyone that can would be appriciated.

 

[UFHH01]

Even that you wrote, that the solution to create a new certificate didn't work for you, I would like to include the whole procedure for this:

cp /opt/psa/var/certificates/XXXXX{,.savedbysupport}
openssl req -x509 -nodes -days 365 -newkey rsa:2048 -keyout /opt/psa/var/certificates/XXXXX -out /opt/psa/var/certificates/XXXXX

/usr/local/psa/admin/bin/httpdmng --reconfigure-all

service apache2 restart && service php5-fpm && service nginx restart && service psa restart

Depending on your operating system, the paths might be slightly different on your system, so it might be "/usr/local/psa/var/certificates/" and instead of "service servicename restart" you might use "/etc/init.d/servicename restart" - please check paths and servicenames before using this solution.

In Plesk version 12, there is as well a pretty nice feature called "Plesk Webserver Configuration Troubleshooter" implemented ( Parallels Knowledge Base: KB 115587 ), you might find it usefull to use it.

 

[GharibK]

did what u said and got this after the reconfigure part

Details: [2014-09-12 14:44:07] ERR [util_exec] proc_close() failed
Can not reload proxy server: Service /etc/init.d/nginx failed to reload
Service /etc/init.d/nginx failed to reload

 

[UFHH01]

Maybe nginx wasn't started, so a command like "restart" doesn't work and will respond with a failure like "failed to reload". Please try "/etc/init.d/nginx stop" "/etc/init.d/nginx start" and see what your system replies.

 

[GharibK]

did that and it replayed:

Starten van nginx: nginx: [emerg] open() "/var/run/nginx.pid" failed (13: Permission denied)
[ OK ]


but the apache server is online now.

now i got a nother problem the http (port 80) is offline

logged on plesk and tried to reconfigure the webserver and got this replay:

Error: Can not reconfigure webserver configurations: Unable to execute httpdmng: Error occured while sending feedback. HTTP code returned: 502
Error occured while sending feedback. HTTP code returned: 502
Execution failed.
Command: httpdmng
Arguments: Array
(
[0] => --reconfigure-server
[1] => -no-restart
)

Details: [2014-09-12 15:41:49] ERR [util_exec] proc_close() failed
Error occured while sending feedback. HTTP code returned: 502
[2014-09-12 15:41:54] ERR [util_exec] proc_close() failed
Error occured while sending feedback. HTTP code returned: 502
[2014-09-12 15:41:55] ERR [panel] Apache config (14105293060.99083400) generation failed: Template_Exception: nginx: [emerg] open() "/var/run/nginx.pid" failed (13: Permission denied)
nginx: configuration file /etc/nginx/nginx.conf test failed

file: /usr/local/psa/admin/plib/Template/Writer/Webserver/Abstract.php
line: 75
code: 0
Error occured while sending feedback. HTTP code returned: 502
nginx: [emerg] open() "/var/run/nginx.pid" failed (13: Permission denied)
nginx: configuration file /etc/nginx/nginx.conf test failed

 

[UFHH01]

Öhm, sorry GharibK, did you use the command

cp /opt/psa/var/certificates/XXXXX{,.savedbysupport}
openssl req -x509 -nodes -days 365 -newkey rsa:2048 -keyout /opt/psa/var/certificates/XXXXX -out /opt/psa/var/certificates/XXXXX

... without changing the XXXXX to the actual certificate - name/number? YOUR specific command, regarding to your previous log would be:

cp /opt/psa/var/certificates/certFATV5bZ{,.savedbysupport}
openssl req -x509 -nodes -days 365 -newkey rsa:2048 -keyout /opt/psa/var/certificates/certFATV5bZ -out /opt/psa/var/certificates/certFATV5bZ

Please ALWAYS replace custom used signs like "XXXXX" or custom used definition like "YOURDOMAIN.com", suggested in threads, posts and KBs with YOUR SPECIFIC definition - otherwise the given solutions never work the way they are suggest for.

 

[GharibK]

i changed it to my certificate so thats not the problem

i need to get the http port 80 online and i think its solved

 

[UFHH01]

What is the ouput, if you stop apache, php5-fpm and nginx and start the services manually again?

STOP SERVICES:

/etc/init.d/apache2 stop
( or instead of apache2 = httpd ?!? )

/etc/init.d/php5-fpm stop
/etc/init.d/nginx stop

START SERVICES:

/etc/init.d/apache2 start
( or instead of apache2 = httpd ?!? )

/etc/init.d/php5-fpm start
/etc/init.d/nginx start

 

[GharibK]

(/etc/init.d/httpd stop/start)

# /etc/init.d/httpd stop
Stoppen van httpd: [ OK ]

]# /etc/init.d/httpd start
Starten van httpd: [ OK ]

(/etc/init.d/php5-fpm stop/start)

# /etc/init.d/php5-fpm stop
-bash: /etc/init.d/php5-fpm: File or directory does not exist

]# /etc/init.d/php5-fpm start
-bash: /etc/init.d/php5-fpm: File or directory does not exist


(/etc/init.d/nginx stop/start)

/etc/init.d/nginx stop (no replay)

# /etc/init.d/nginx start
Starten van nginx: nginx: [emerg] open() "/var/run/nginx.pid" failed (13: Permission denied)
[ OK ]


this is what i got

 

[UFHH01]

That's why I suggested to control the paths BEFORE just copying the solutions to the shell command line. ^^ Did you use FASTCGI before upgrading?

First of all you should be informed, that Plesk installed a proxy server called "nginx", which you either use or if you don't then please uninstall that package now, because you should inform yourself before using it. Your domain configuration will be completly changed ( that's why you have a tool like "httpdmng" to reconfigure your domain configurations ) and some modifications have to be made, to speed up http requests with nginx.

You may install or uninstall every included Plesk - Packages over the URL: https://YOURDOMAIN.COM:8447/select_components.html.
The NGINX web server is a feature under "Web hosting features" and is an option, which you might choose or not. - If you never heard of it, and don't want to learn something about the features and options, you would probably choose to leave this packages uninstalled.

The same with FASTCGI ( Apache mod_fcgid module ). If you never heard of it and don't know in the first place how to use it, how to configure it and what ever else, please uninstall the package and inform yourself about advantages or disadvantages, before you install this package.

After uninstalling nginx, make sure, that you run "/usr/local/psa/admin/bin/httpdmng --reconfigure-all" again, because your webserver needs different domain configurations without nginx.

After re-configuring the http-conf's, you should restart your webserver apache again with YOUR command: /etc/init.d/httpd start


... now your domains should be accessed the "normal" way, that you already know. If you still would like to configure the AddOn "NGINX" and "FASTCGI", please reply again with a different thread so someone may guide you and inform you about modifications and configurations.

 

[GharibK]

thank you UFHH01 im really happy now my server is working good again and i wil start a new thread for nginx and fastcgi when im ready for it becaus i dont need it at the moment.