Zayatzz's questions

I have a following wordpress nginx configuration in my /etc/nginx/sites-available/wordpress:

    index index.html index.htm index.nginx-debian.html;

    location / {
        proxy_pass http://nextjs;
        proxy_http_version 1.1;
            proxy_set_header X-Forwarded-Host $host;
            proxy_set_header X-Forwarded-Server $host;
            proxy_set_header X-Real-IP $remote_addr;
            proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
            proxy_set_header X-Forwarded-Proto $scheme;
            proxy_set_header Host $http_host;
            proxy_set_header Upgrade $http_upgrade;
            proxy_set_header Connection "Upgrade";
            proxy_pass_request_headers on;
    }


    location /wordpress {
        root /var/www;
        try_files $uri $uri/wordpress/ /wordpress/index.php$is_args?$args;

            location ~ \.php$ {
            include fastcgi_params;
            fastcgi_pass unix:/run/php/php8.1-fpm.sock;
            fastcgi_split_path_info ^(.+\.php)(/.+)$;
            fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
                    include snippets/fastcgi-php.conf;
                    fastcgi_buffers 1024 4k;
                    fastcgi_buffer_size 128k;
            }
    }

As you can seem, there is a static, NextJS-based frontend for WordPress. And generally, it works just great. There is this one quirk that does not work, though, and I'm not sure if it is a WordPress issue or a Nginx issue.

• When the user navigates to /wordpress, then we see content from wordpress, which is fine
• When the user navigates to /wordpress/graphql, then the grahpql endpoint is working just fine
• When the user navigates to /wordpress/wp-admin, then the things break. There is a redirect loop, and the brauser reports that the site is not redirecting properly and so on.
• when the user navigates to /wordpress/wp-admin/index.php, then everything works fine though.

So where does this come from? Is it a Nginx thing? WordPress thing? How do I fix this?

Edit 1: Added curl -vvv output:

C:\Users\alank>curl -vvv https://mydomain.ee/wordpress/wp-admin
*   Trying 194.204.13.171:443...
* Connected to mydomain.ee (194.204.13.171) port 443 (#0)
* schannel: disabled automatic use of client certificate
* ALPN: offers http/1.1
* ALPN: server did not agree on a protocol. Uses default.
* using HTTP/1.x
> GET /wordpress/wp-admin HTTP/1.1
> Host: mydomain.ee
> User-Agent: curl/8.0.1
> Accept: */*
>
* schannel: remote party requests renegotiation
* schannel: renegotiating SSL/TLS connection
* schannel: SSL/TLS connection renegotiated
* schannel: remote party requests renegotiation
* schannel: renegotiating SSL/TLS connection
* schannel: SSL/TLS connection renegotiated
< HTTP/1.1 302 Found
< server: nginx/1.18.0 (Ubuntu)
< date: Tue, 22 Aug 2023 10:19:10 GMT
< content-type: text/html; charset=UTF-8
< transfer-encoding: chunked
< expires: Wed, 11 Jan 1984 05:00:00 GMT
< cache-control: no-cache, must-revalidate, max-age=0
< link: <https://mydomain.ee/wordpress/wp-json/>; rel="https://api.w.org/"
< x-redirect-by: WordPress
< location: https://mydomain.ee/wordpress/wp-admin/
<
* Connection #0 to host mydomain.ee left intact

Edit 2: curl -vvv with trailing slash:

C:\Users\alank>curl -vvv https://mydomain.ee/wordpress/wp-admin/
*   Trying 194.204.13.171:443...
* Connected to  mydomain.ee (194.204.13.171) port 443 (#0)
* schannel: disabled automatic use of client certificate
* ALPN: offers http/1.1
* ALPN: server did not agree on a protocol. Uses default.
* using HTTP/1.x
> GET /wordpress/wp-admin/ HTTP/1.1
> Host: mydomain.ee
> User-Agent: curl/8.0.1
> Accept: */*
>
* schannel: remote party requests renegotiation
* schannel: renegotiating SSL/TLS connection
* schannel: SSL/TLS connection renegotiated
* schannel: remote party requests renegotiation
* schannel: renegotiating SSL/TLS connection
* schannel: SSL/TLS connection renegotiated
< HTTP/1.1 302 Found
< server: nginx/1.18.0 (Ubuntu)
< date: Tue, 22 Aug 2023 11:13:01 GMT
< content-type: text/html; charset=UTF-8
< transfer-encoding: chunked
< expires: Wed, 11 Jan 1984 05:00:00 GMT
< cache-control: no-cache, must-revalidate, max-age=0
< link: <https://mydomain.ee/wordpress/wp-json/>; rel="https://api.w.org/"
< x-redirect-by: WordPress
< location: https://mydomain.ee/wordpress/wp-admin/
< 
* Connection #0 to host mydomain.ee left intact

I have a problem with my current setup... And the issue is, that if i set DEBUG = False in django settings.py file nginx stops showing django tracebacks on 500 errors, but does not show our 500 page either. It just displays nginx 502 bad gateway error.

I do get the email error with traceback, just as i've configured it to mail tracebacks to me, when they occur. But i want to display nice 500 page to users not nginx 502 gateway error...

Honestly, i dont even know where to begin searching the root of the problem. Im prepared to post all necessary config files, if some nginx expert would come and tell me what he wants to see.

alan

Edit1: I looked up what log file shows at one of those 500 errors, and it shows this:

[pid: 16203|app: 0|req: 1/1] my.ip.address () {46 vars in 915 bytes} [Thu Sep 12 10:01:17 2013] GET /settings/personal/ => generated 0 bytes in 1249 msecs (HTTP/1.1 500) 0 headers in 0 bytes (0 switches on core 0)

Does this mean that its somehow django fault cause it looks like django returned 0 bytes?

I just did clean wordpress install and set up its .htaccess like they say i should do :

http://codex.wordpress.org/Using_Permalinks

# BEGIN WordPress
<IfModule mod_rewrite.c>
RewriteEngine On
RewriteBase /
RewriteRule ^index\.php$ - [L]
RewriteCond %{REQUEST_FILENAME} !-f
RewriteCond %{REQUEST_FILENAME} !-d
RewriteRule . /index.php [L]
</IfModule>
# END WordPress

when i try to go to some address in my browser like http://localhost:888/blog/ i get 404 error from server. I dont even get directed by wordpress to wordpess's own 404 page.

When i look up my site's error log:

[Sat Apr 20 10:43:36 2013] [error] [client 127.0.0.1] File does not exist: /home/alan/projects/pin_wp/blog, referer: http://localhost:888/

So this leads me into thinking that i have messed up my virtualhost conf somewhere - since im one big noob in that area:

alan@alan:~/projects/pin_wp$ cat /etc/apache2/sites-enabled/localhost-pinwp 
Listen 888
<VirtualHost *:888>
    ServerAdmin webmaster@localhost

    DocumentRoot /home/alan/projects/pin_wp
    ServerName html

    <Directory />
        Options FollowSymLinks
        AllowOverride None
    </Directory>
    <Directory /home/alan/projects/pin_wp>
        Options Indexes FollowSymLinks MultiViews +Includes
        AllowOverride None
        Order allow,deny
        allow from all
        AddOutputFilter INCLUDES .html
    </Directory>

    ScriptAlias /cgi-bin/ /usr/lib/cgi-bin/
    <Directory "/usr/lib/cgi-bin">
        AllowOverride None
        Options +ExecCGI -MultiViews +SymLinksIfOwnerMatch
        Order allow,deny
        Allow from all
    </Directory>

    ErrorLog ${APACHE_LOG_DIR}/error-pinwp.log

    # Possible values include: debug, info, notice, warn, error, crit,
    # alert, emerg.
    LogLevel warn

    CustomLog ${APACHE_LOG_DIR}/access-pinwp.log combined

    Alias /doc/ "/usr/share/doc/"
    <Directory "/usr/share/doc/">
        Options Indexes MultiViews FollowSymLinks
        AllowOverride None
        Order deny,allow
        Deny from all
        Allow from 127.0.0.0/255.0.0.0 ::1/128
    </Directory>

</VirtualHost>

Could someone please help me fix this issue and point me out what kind of dumb mistakes i've done this time.

Alan

I needed to install UCS2 python next to UCS4 python. So I went to comp.lang.python and asked them about it. Probably not the best place to ask it, but they answered

https://groups.google.com/forum/?fromgroups#!topic/comp.lang.python/bGuAfqa76W8

and now i have brand new python 2.7.3 ucs2 installed in /opt/bin/python

What I need now is - how can I install all other python modules that I have installed for that python version also. Basically stuff like PIL and postgresql and mod_wsgi - basically everything needed to run Django for that python version.

Is this the right the place to ask for it?

Im trying to build application, which uses 2 mysql servers. I set up port forwarding (thats the name to callit -i dont know) like that:

sudo ssh -g -L3306:foreignserver.ip:3306 [email protected]

and i changed my own my.cnf port to 3360 and did

sudo service mysql restart

When i try to connect to the foreignserver - i fail.

mysql --host=localhost --port=3306 --user=userinforeignsystem --password=pwdthere

gives me error

ERROR 1045 (28000): Access denied for user 'userinforeignsystem'@'localhost' (using password: YES)

since i cant connect to the other server mysql database from outside, ssh/port forwarding is only way to develop for me. Database user privileges and usernames/passwords are doublechecked and correct. I could log in, using those credentials, in other server just fine.

Alan.

Edit: If i swap the username and password for my own mysql username and password and leave the port 3306, it somehow connects to my own computers mysql - i can select databases and see tables of my own mysql db not foreignserver.ip stuff. It looks like it either pays no aatention to port or something is wrong with port forwarding or something...

Edit2:

After some googling i found command

sudo netstat -plntu

and used it. When i removed lines not including ports 3360 and 3306 only those 2 were left:

tcp        0      0 0.0.0.0:3306            0.0.0.0:*               LISTEN      8145/mysqld     
tcp6       0      0 :::3306                 :::*                    LISTEN      8080/ssh  

Does that mean, that for some reason, my mysql is not running at 3360 despite the my.cnf?

Edit3: localhost/phpmyadmin variables also shows port 3306 in use.. guess thats where the problem lies...

Edit4: same time mysqladmin variables command shows port beeing 3360. Now im out of ideas.

edit5: doublechecked my.cnf and found out other places where port was set:

Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name
tcp        0      0 0.0.0.0:3360            0.0.0.0:*               LISTEN      -               
tcp        0      0 0.0.0.0:3306            0.0.0.0:*               LISTEN      -               
tcp6       0      0 :::3306                 :::*                    LISTEN      -               

and now phpmyadmin shows port 3360 beeing used too. The mysql command from shell still fails to connect though.

EDIT6 - SOLVED. For some reason

mysql --host=localhost --port=3306 --user=userinforeignsystem --password=pwdthere

did not work, while

mysql --host=127.0.0.1 --port=3306 --user=userinforeignsystem --password=pwdthere

worked. If someone could explain that to me, then youd be teaching me something interesting and new :)

At one hosting company, they used to run python projects with fcgi. They had set it up so that when i changed django.fcgi file, which put django & my project on pythonpath, my project code was instantly recompiled.

Because of that a friend set up hosting for our shared project in his server using fastcgi. It has been set up and the python scripts execute as they should, but what we do not know is, how to set it up so that my project would be recompiled when my setup file has been changed.

Alan

Hell0

We (our company) has been using one CMS for several websites that we have created. We have encountered this particular problem several times now and we cant think of a fix for it.

Here's the description of the problem: In this cms there is possible to create picture galleries. When you create folder for gallery via CMS then folder is beeing created by www user and its permissions are set to 0666. When user uploads pictures to this gallery using its editor view, then the pictures are uploaded to previously created folder and cut into necessary sizes. .thumbnails and other folders are created to hold pictures of different sizes.

If you create folder for gallery via ftp, then the owner of the gallery is ftp user. If you manually set permissions to 0666 and upload the files with correct permissions, everything seems to be fine. Now in the public view, if you choose this previously created folder as your gallery folder, the cms fails to create .thumbnail subdirectory (and other required subdirectories also). The only difference between the folders is the owner of the folders.

Right now i am trying to get my hands on this hosting accounts php log. I guess i will have more information then.

Alan

I do not know alot about linux so please take a bit longer time trying to figure out what i am asking for.

background story can be read here : http://groups.google.com/group/django-users/browse_thread/thread/5cd35edc04876523

But here is short recap: I installed stackless python and found that i have to compile myself several python libraries to get my django project up and running again. I compiled setuptools & python-mysqldb, I downloaded and compiled jpeg, freetype2 and PIL.

Right now the problem is that django fails to validate imagefile when i upload file with a modelform. Do not turn this question down if you know nothing about django or python. this question is not neccessarily all about that.

When tracking down the point in django code which raises the error i found that it was done at the point where python used Image.load() function on uploaded image. When i commented that line out i recieved this error:

IOError at /profile/edit/

decoder jpeg not available

Request Method:     POST
Request URL:    127.0.0.1/profile/edit/
Exception Type:     IOError
Exception Value:    decoder jpeg not available
Exception Location:     /usr/local/lib/python2.6/site-packages/PIL/Image.py in _getdecoder, line 375
Python Executable:  /usr/local/bin/python
Python Version:     2.6.2
Python Path:    ['/usr/local/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg', '/usr/local/lib/python2.6/site-packages/MySQL_python-1.2.3c1-py2.6-linux-i686.egg', '/usr/local/lib/python2.6', '/usr/local/lib/python2.6/plat-linux2', '/usr/local/lib/python2.6/lib-tk', '/usr/local/lib/python2.6/lib-old', '/usr/local/lib/python2.6/lib-dynload', '/usr/local/lib/python2.6/site-packages', '/usr/local/lib/python2.6/site-packages/PIL', '/home/projects', '/home/projects/tst']

As i understand it fails to do something with jpeg. Could that mean that even though i have compiled and installed this jpeg library ( jpegsrc.v7.tar.gz from ww.ijg.org), there is still something wrong with paths or something else.

I did try doing sudo make uninstall in the folder where i had compiled jpeg and it removed the files that I had installed. But ubuntu installation also had this library installed before i ever did anything with stackless.

So after all this long story the question is: How can i get rid of Jpeg from my computer so i can make sure that jpeg decoder used everywhere is the jpeg decoder that i have compiled and installed?

Alan