Upgrading to 11 works, but Nginx crashed Plesk

[Elfchen]

I have updated to V 11.0.9.

The update was well and everything worked.

After that I decided to install nginx.

Since I installed it no domain is accessable.

All domains point to:

It works!

This is the default web page for this server.

The web server software is running but no content has been added, yet.

I can't find any way to remove nginx to get Plesk working again.

All vhosts are configurated to 127.0.0.1 for nginx usage and nginx is running, but it seems, that somewhere it did not read configuration files, but I can't find where.

When I try to restart nginx I get:

Fehler: Unable to make action: Unable to manage service by nginx_control: /opt/psa/admin/sbin/nginx-config execution failed:
nginx: [warn] conflicting server name "www.domain.com" on 1.2.3.4:80, ignored

... repeated for ALL domains ...

till end of output:
nginx: [emerg] open() "/var/www/vhosts/www.domain9999com/statistics/logs/proxy_access_log" failed (24: Too many open files)
nginx: configuration file /etc/nginx/nginx.conf test failed ('--restart', 'nginx')

Increasing of ulimit will solve last error (24), but all the others still exists and I did not find where and why.

For all websites I receive only a 404 error when I try to access them!

When I now click on stop nginx I get:

Fehler: Unable to make action: Unable to manage service by nginxmng: #ATTENTION!
#
#DO NOT MODIFY THIS FILE BECAUSE IT WAS GENERATED AUTOMATICALLY,
#SO ALL YOUR CHANGES WILL BE LOST THE NEXT TIME THE FILE IS GENERATED.
#
#IF YOU REQUIRE TO APPLY CUSTOM MODIFICATIONS, PERFORM THEM IN THE FOLLOWING FILES:

#/var/www/vhosts/mysqldumper..........

Now when I try to access websites it did not get correct site from vhost path, it displays content of /var/www - Apache default page with: It works ! - funny - nothing works here!

Back in Plesk panel I get:

Aufgrund von Fehlern in den Konfigurations-Templates konnten keine neuen Konfigurationsdateien für den Apache Webserver erstellt werden: Configuration changes scheduled and not applied. Eine detaillierte Fehlernachricht wurde Ihnen per E-Mail zugesendet. Schauen Sie bitte in Ihr Postfach, beheben Sie die Fehler und klicken Sie hier, um die fehlgeschlagene Konfiguration erneut zu erstellen oder klicken Sie hier, um alle Konfigurationen erneut zu erstellen.

So click on - write new configuration ...

Die Neukonfigurierung des Webservers ist in Bearbeitung. Verlassen bitte diese Seite nicht, bis der Vorgang abgeschlossen ist.

That returns no error:

Die Neukonfigurierung des Webservers ist abgeschlossen: Die neuen Konfigurationsdateien wurden erfolgreich erstellt und übernommen.

But all domains now point to /var/www and displaying default apache site "it works!"


Can anyone help to get server back running?

 

[Blake@Parallels]

You can try the following to regenerate the vhost configurations:
# /usr/local/psa/admin/sbin/httpdmng --reconfigure-all

 

[abdi]

Nginx Related

Personally I have note been succesful with nginx that comes with plesk. When I installed it, 40% of the domains were showing 1 particular website which is not even the default for the IP address.

While I still wait for a clearer and valid solution, do what I did:

Just go to the control panel -> settings -> services -> and switch off Nginx :
And if you want it on again, switch it on again from same place.

Alternatively you can run the commands below:

/usr/local/psa/admin/bin/nginxmng --disable
/usr/local/psa/admin/bin/nginxmng --enable

 

[TorstenD]

in my Plesk version crashes the Navigation on the left side

ERROR: Exception
[unixODBC][MySQL][ODBC 3.51 Driver]Access denied for user 'apsc'@'localhost' (using password: YES)

Additionally, an exception has occurred while trying to report this error: Exception
[unixODBC][MySQL][ODBC 3.51 Driver]Access denied for user 'apsc'@'localhost' (using password: YES)

0: aps_php.php:8075
aps_controller_registry_getpackage(resource, string '2f30f91b-cd0d-4f5a-9446-709de3e09683')
1: aps_php.php:8075
APS_Controller_Registry->getPackage(string '2f30f91b-cd0d-4f5a-9446-709de3e09683')
2: Config.php:21
HelpDesk_Config::getHelpDeskInstances()
3: NavigationContext.php:342
NavigationContext->_fetchHelpDeskSectionNodes(object of type UserAdmin)
4: NavigationContext.php:196
NavigationContext->fetchCustomButtonsSectionNodes(object of type UserAdmin)
5: NavigationContext.php:70
NavigationContext->fetchSection(object of type UserAdmin, string 'custom_buttons', array)
6: NavigationContext.php:402
NavigationContext->get(object of type UserAdmin)
7: Navigation.php:557
Navigation->getContextMenu(object of type UserAdmin, object of type Skin)
8: left.php3:4

some ideas?

 

[abdi]

Looks like database of APS controller was not created during upgrade procedure for some reasons. I suggest you contact Support Team. They will correctly create this database for you and fix this issue.

Details about that issue also answered on the forum:

http://forum.parallels.com/showthread.php?t=258935

 

[TorstenD]

Thx abdi, but i hope someone here can help me.

After the update everything works fine then i installed ngix as additional component. After install the Navigation crashes

I use:

# cat /usr/local/psa/version
11.0.9 CentOS 5 110120608.16


# mysql -uadmin -p`cat /etc/psa/.psa.shadow` -D psa -e 'select * from upgrade_history order by upg rade_date desc limit 1'
+---------------------+--------------+------------+
| upgrade_date | version_info | db_version |
+---------------------+--------------+------------+
| 2012-06-24 20:35:52 | 11.0.9 | 11.0.9 |
+---------------------+--------------+------------+


I updated to: psa-11.0.9-cos5.build110120608.16

in the attached Logfile is somthing about Sitebuilder and Billing, but i`ll never use this.

 

[abdi]

Good thing is since you dont plan to use plesk billing, start by removing it with the commands:

rpm -e sw-sso plesk-billing plesk-billing-plugins
/etc/init.d/psa restart

Then after words, if the problem still persists then:

Run the autoinstaller and install about anything in list of applications you may not have installed, eg QMail instead of the Postifix or the reverse (Ofcourse you can reverse later)

Doing that above will prompt plesk to rebuild the APS database and resolve un-finished issues. Let me know how it goes ..

 

[TorstenD]

hi,

i found in my logfiles a part that i have not enough disc space in /usr

i give them 2gb more and restart psa to see whats happend. the autoupdate process is now still working and i must wait a while.

i hope this will fix it...

 

[abdi]

Alright, just let me know how it goes.

 

[TorstenD]

now i will reconfigure Plesk, but i don`t find a command line solution. i search for a command that all installed components will reinstall or find a command to deinstall only ngix.

any ideas?

 

[abdi]

yum remove sw-nginx

That will remove only nginx

 

[TorstenD]

ok...

i remove the ngix and reinstall with plesk, but the error is still available...

i think i must reinstall the complete plesk 11 version... or find a solution with the database

...what a sunday...

 

[abdi]

Just as a by the way, but sometimes can also help:

Check that the MySQL server is accepting connections via TCP IP:

#mysql -h 127.0.0.1 -P 3306 -uadmin -p`cat /etc/psa/.psa.shadow`

The MySQL server may not be accepting connections via TCP IP because of your firewall settings, network link or MySQL server settings.

For the MySQL server settings, you can check the /etc/my.cnf folder for the "skip-networking" parameter. If this parameter is present, then the MySQL server will not accept network connections via TCP.

To enable TCP connections, comment on the "skip-networking" parameter by adding the "#" character ("#skip-networking"). After adding the character, restart the MySQL server by the following command:

# /etc/init.d/mysqld restart

 

[TorstenD]

This works...

mysql -h 127.0.0.1 -P 3306 -uadmin -p`cat /etc/psa/.psa.shadow`
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 1429
Server version: 5.0.95 Source distribution

Copyright (c) 2000, 2011, Oracle and/or its affiliates. All rights reserved.

Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

mysql>

i think this was the first error in /var/log/sw-cp-server/error_log

2012-06-24 20:25:54: (mod_fastcgi.c.2746) FastCGI-stderr: Exception: [unixODBC][MySQL][ODBC 3.51 Driver]Access denied for user 'apsc'@'localhost' (using password: YES)
file: /usr/share/aps_php/aps_php.php
line: 8075
code: 0
trace: #0 /usr/share/aps_php/aps_php.php(8075): aps_controller_registry_getpackage(Resource id #149, '2f30f91b-cd0d-4...')
#1 /usr/local/psa/admin/plib/HelpDesk/Config.php(21): APS_Controller_Registry->getPackage('2f30f91b-cd0d-4...')
#2 /usr/local/psa/admin/plib/NavigationContext.php(342): HelpDesk_Config::getHelpDeskInstances()
#3 /usr/local/psa/admin/plib/NavigationContext.php(196): NavigationContext->_fetchHelpDeskSectionNodes(Object(UserAdmin))
#4 /usr/local/psa/admin/plib/NavigationContext.php(70): NavigationContext->fetchCustomButtonsSectionNodes(Object(UserAdmin))
#5 /usr/local/psa/admin/plib/NavigationContext.php(402): NavigationContext->fetchSection(Object(UserAdmin), 'custom_buttons', Array)
#6 /usr/local/psa/admin/plib/Navigation.php(557): NavigationContext->get(Object(UserAdmin))
#7 /usr/local/psa/admin/htdocs/left.php3(4): Navigation->getContextMenu(Object(UserAdmin), Object(Skin))
#8 {main}

 

[TorstenD]

i found this: http://forum.parallels.com/printthread.php?t=111989

and i try the bootstrapper.sh repair solution, but it does not work for plesk and the aps...

 

[TorstenD]

i found a other solution, but who can send me a database dump?

http://forum.parallels.com/pda/index.php/t-246791.html

 

[TorstenD]

i try to login to the user GUI, but still the same problem

 

[IgorG]

Normally this type of issue occur due to the password mismatch between "misc" table in psa database and "user" table in mysql database.

In order to fix the issue, try to reset the password in mysql database 'user' table with something like:

mysql> select * from psa.misc where param like "%aps%";
+--------------+--------------+
| param | val |
+--------------+--------------+
| aps_database | apsc |
| aps_host | localhost |
| aps_login | apsc |
| aps_password | t**********P |
| aps_port | 3306 |
+--------------+--------------+
5 rows in set (0.00 sec)

mysql> update mysql.user set password=PASSWORD('t*********P') where User='apsc';

After that run bootstrapper repair procedure.

 

[Blake@Parallels]

Normally this type of issue occur due to the password mismatch between "misc" table in psa database and "user" table in mysql database.

In order to fix the issue, try to reset the password in mysql database 'user' table with something like:

mysql> select * from psa.misc where param like "%aps%";
+--------------+--------------+
| param | val |
+--------------+--------------+
| aps_database | apsc |
| aps_host | localhost |
| aps_login | apsc |
| aps_password | t**********P |
| aps_port | 3306 |
+--------------+--------------+
5 rows in set (0.00 sec)

mysql> update mysql.user set password=PASSWORD('t*********P') where User='apsc';

After that run bootstrapper repair procedure.

To be clear, the password to use in single quotes in the mySQL 'update' command should be the same as returned from the previous 'select' query.

After this, the bootstrapper command line is:

# /usr/local/psa/bootstrapper/pp11.0.9-bootstrapper/bootstrapper.sh repair

 

[abdi]

Torsten, please let us know if that worked for you ...