Just recently moved a domain from another host to my newly setup Debian Wheezy nginx setup on a VPS. For some reason I kept getting error 500 trying to load a page no matter what I did. Checked the logs and saw Code: FastCGI sent in stderr PHP Parse error: syntax error, unexpected 'array' (T_ARRAY) while reading response header from upstream /var/www/clients/client2/web1/web/wp-includes/formatting.php on line 1691 Line 1691 in the formatting.php file is pretty short: Code: return $matches[0]; After trying all the suggested site moving fixes I could find, like changing the .htaccess file, disabling plugins & themes, etc. - it occured to me that I had already moved another site that was using Wordpress and it had worked without any issues. Further investigation revealed the working site was using Wordpress 3.8.3. So I downgraded the non working site from 3.9 to 3.8.3 and it worked instantly. So the question is - Why did 3.9 break? PHP is on PHP Version 5.4.4-14+deb7u9 MySQL version is 5.5.37-0+wheezy1 Any ideas? Has anybody else seen this issue yet?
Just to make it even stranger. I tired doing a new wordpress install with version 3.8.3 on a different site. I used the same files that somehow magically fixed the existing site. All I get is an error 500. And in the error log for the site ( /var/log/ispconfig/domain ) here is what I get. Code: 2014/05/03 15:31:23 [error] 15696#0: *19650 FastCGI sent in stderr: "PHP message: PHP Parse error: syntax error, unexpected 'to' (T_STRING) in /var/www/clients/client2/web2/web/wp-includes/taxonomy.php on line 2060" while reading response header from upstream, client: bbb.aaa.xxx.yyy, server: domain.com, request: "GET /wp-admin/install.php HTTP/1.0", upstream: "fastcgi://unix:/var/lib/php5-fpm/web2.sock:", host: "domain.com" (Domain name and IP address have been changed)
Here are my Nginx directives Code: client_max_body_size 100M; location / { try_files $uri $uri/ /index.php?$args; } location /roundcube { root /var/lib/; index index.php index.html index.htm; location ~ (.+\.php)$ { try_files $uri =404; include /etc/nginx/fastcgi_params; # To access SquirrelMail, the default user (like www-data on Debian/Ubuntu) must be used #fastcgi_pass 127.0.0.1:9000; fastcgi_pass unix:/var/run/php5-fpm.sock; fastcgi_index index.php; fastcgi_intercept_errors on; fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name; fastcgi_buffer_size 128k; fastcgi_buffers 256 4k; fastcgi_busy_buffers_size 256k; fastcgi_temp_file_write_size 256k; } location ~* /.svn/ { deny all; } location ~* /README|INSTALL|LICENSE|SQL|bin|CHANGELOG$ { deny all; } } location /webmail { rewrite ^ /roundcube last; } location /phpmyadmin { root /usr/share/; index index.php index.html index.htm; location ~ ^/phpmyadmin/(.+\.php)$ { try_files $uri =404; root /usr/share/; fastcgi_pass 127.0.0.1:9000; fastcgi_param HTTPS on; # <-- add this line fastcgi_index index.php; fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name; include /etc/nginx/fastcgi_params; fastcgi_buffer_size 128k; fastcgi_buffers 256 4k; fastcgi_busy_buffers_size 256k; fastcgi_temp_file_write_size 256k; fastcgi_intercept_errors on; } location ~* ^/phpmyadmin/(.+\.(jpg|jpeg|gif|css|png|js|ico|html|xml|txt))$ { root /usr/share/; } } location /phpMyAdmin { rewrite ^/* /phpmyadmin last; }
Just installed latest wordpress 3.9 from wordpress.org and it works ok. It seems that its something about the update that makes you problems. Also Im using mod-php not suphp.
It's just been so inconsistent so far on how it reacts to WP. like it worked fine moving over a WP 3.8.3 site, but not moving 3.9 using the same procedure. ( Import database, move files, update config ) And yesterday got errors trying to create a brand new 3.8.3 site with a blank database. Then just for grins I decided to try it with the 3.9 - and it worked! I don't mind if there are things that will or will not work on my system as long as I know what they are, and why they won't work so I can explain that to customers. I did at some point during this adventure restart both nginx and php5-fpm. But did not make any configuration changes to either. Code: service php5-fpm restart service nginx restart Scratching my head on this one a bit. For now I'm just going to ignore it and move on, I guess. It's a low volume server, 10 or so sites. Only 2 of which are running Wordpress and one of those is a personal site.