Apache2 problem after upgrading from Wheezy LTS to Jessie LTS

  • Hello all,


    I upgraded my vserver with below commands

    • Update your /etc/apt/sources.list according to the list above.
    • Update your system. apt-get update
    • Install apt from Jessie before upgrading your system. apt-get install apt -t jessie
    • Upgrade your system. apt upgrade
    • Upgrade your system and remove obsolete packages. apt dist-upgrade
    • Carefully check all debconf prompts and update configuration files as needed.

    But now i have problem with apache2 and upgrading froxlor


    /etc/init.d/apache2 restart

    [....] Restarting web server: apache2AH00548: NameVirtualHost has no effect and will be removed in the next release /etc/apache2/sites-enabled/10_froxlor_ipandport_xx.xx.170.228.80.conf:5


    And now i cannot access my webpages :(

    Please help

  • Already gone worst :(

    /etc/init.d/apache2 restart

    [FAIL] Restarting web server: apache2 failed!

    [warn] The apache2 configtest failed. ... (warning).

    Output of config test was:

    AH00548: NameVirtualHost has no effect and will be removed in the next release /etc/apache2/sites-enabled/10_froxlor_ipandport_xx.xx.170.228.443.conf:5

    AH00526: Syntax error on line 11 of /etc/apache2/sites-enabled/28_froxlor_normal_vhost_xxx.v22xxxx.yourvserver.net.conf:

    Invalid command 'php_admin_value', perhaps misspelled or defined by a module not included in the server configuration

    Action 'configtest' failed.

    The Apache error log may have more information.

  • Either your Froxlor really sets and invalid php_admin_value or your Apache2 is missing a php module (apache2-mod-php*).

    A wrong php_admin_value might also be the result of a manual entry for a vhost.


    In that case, since probably your Froxlor by now is unaccessible, you could do the following to regain control:

    1. stop the cron daemon.

    2. backup your apache2 configuration directory first - especially all „froxlor-sites“ unless the file are contained within „sites-enabled“.

    3. then delete all configuration files created by froxlor for you vhost („froxlor-sites“),

    4. restart the apache2 service, which now only brings up the default host.

    5. Resolve the error within Froxlor's GUI.

    6. then recreate the config files from within froxlor and

    7. manually invoke froxlor's master cronjob at the console with the --force option and see, if apache is now coming up without problems - if not repeat the procedure until the error is resolved.

  • Zitat

    Invalid command 'php_admin_value', perhaps misspelled or defined by a module not included in the server configuration

    Are you using mod_php and have you installed it?


    The entry "php_admin_value" is not supported in the .conf file. To test comment this line out and start apache2 again.

  • Thanks eripek
    I deleted apache2 with purge command and deleted all config files


    Installed back apache2 and now I cannot access froxlor :)

    It shows php codes

    Apache2/sites-available/000-default.conf as below;


  • There still must be an error in your apache2 configs.

    What is the output of "apache2ctl configtest" now?


    You can always regain access to froxlor the way I described it, as long as mod-php and the handlers for php are configured correctly.


    Please also have a look at manuals like the following - it is written in german language, but you'll understand the important steps anyways: https://www.allerstorfer.at/fr…-debian-9-x-installieren/

  • eripek


    It says

    Code
    # apache2ctl configtest                                 
    Syntax OK

    Now all ok and php executes again :) Problem caused old wheezy deb paths in /etc/apt and updated new deb


    I checked that website and I screw it :) I used FCGID settings, installed and now I cannot access froxlor again


  • Since Froxlor is storing it's settings in a Database, you can manually revert the settings, set a flag for the cronjob and rerun the master_cronjob over again.


    mysql -u root -p froxlor


    Query: select * from panel_settings where varname like "%fcgi%";


    You will have "1" instead of "0" as a result of this query for "mod_fcgid".


    Change that back: Update set value='1' where varname='mod_fcgid'


    Then (set flag to recreate all configs):

    INSERT INTO `panel_tasks` SET `type`='1';


    and

    php /var/www/froxlor/scripts/froxlor_master_cronjob.php --force


    This would deactivate fcgid again, which I believe would that cause a fallback to mod_php. You can start over from here again. I'd prefer to activate php-fpm, but this can be very tricky at the beginning. Check, that libnss-extrausers is configured well before doing that. Otherwise your fpm-sockets will have the wrong permissions, which leads to a bad gateway Error 500. If that happens, you can always set new permissions on the socket, but the cronjob -unless it is stopped again- will override them later. Permissions will be correct, if libnss-extrausers is well configured.

  • Sorry I haven't been able to check back on this topic earlier.

    I made an error in my previous posting, but I guess you've already figured that out:

    UPDATE panel_settings SET value='1' WHERE varname='mod_fcgid'
    INSERT INTO `panel_tasks` SET `type`='1';

    Still has error

    UPDATE panel_settings SET value='1' WHERE varname='mod_fcgid'

    -> INSERT INTO `panel_tasks` SET `type`='1';

    ERROR 1064 (42000): You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'INSERT INTO `panel_tasks` SET `type`='1'' at line 2