Wordpress will not upload

[priestjim]

Hi all,

I've been getting a strange error with every website running Wordpress ever since I recompiled PHP 5.2.17 with the LSAPI present on LSWS 4.1.11. More specifically the asynchronous file upload script will not work (/wp-admin/async-upload.php) and the admin will get an

HTTP Error

error box. Specifications follow:

OS: Centos 5.8 x86_64
PHP: 5.2.17 / LSAPI
LSWS: 4.1.11

On 2-3 test customers I troubleshooted, the error occurred even with all the plugins turn off. Wordpress was at the latest version. The error occurs only sporadically but it is an issue. The error does not occur on an Apache installation.

Any help would be appreciated!

 

[NiteWave]

any message in error.log and stderr.log ?

 

[priestjim]

Sadly no, these are the clues I have. Even strace'ing the process running the upload shows nothing (to my not that much trained eye).

 

[NiteWave]

please try 4.1.12 to see if any difference. it won't hurt anything --- switching between versions is easy.

 

[priestjim]

Actually, it will. 4.1.12 breaks randomly static file transfers (like images, css or js files) and corrupts them, leaving the client with half-downloaded files.

 

[NiteWave]

it has been identified and fixed in latest build, quite a few days ago. please force-reinstall 4.1.12.

 

[priestjim]

Did the force reinstall, the old issue was fixed but the problem with the uploads remains.

 

[NiteWave]

I tried to reproduce your issue on latest wp-3.3.2 + lsws 4.1.12 + php 5.2.11 at our lab, but failed. using "multi-file uploader".
can you suggest how to reproduce the issue quickly?

 

[priestjim]

Here is some more information about my setup:

- lsphp 5.2.17 with configure flags:

'--disable-debug' '--with-pic' '--disable-rpath' '--without-pear' '--with-bz2' '--with-freetype-dir=/usr' '--with-png-dir=/usr' '--with-xpm-dir=/usr' '--enable-gd-native-ttf' '--with-t1lib=/usr' '--without-gdbm' '--with-gettext' '--with-gmp' '--enable-libevent' '--with-iconv' '--with-jpeg-dir=/usr' '--with-openssl' '--with-pcre-regex=/usr' '--with-zlib' '--with-layout=GNU' '--enable-exif' '--enable-ftp' '--enable-magic-quotes' '--enable-sockets' '--with-kerberos' '--enable-ucd-snmp-hack' '--enable-shmop' '--enable-calendar' '--without-mime-magic' '--with-libxml-dir=/usr' '--with-xml' '--with-system-tzdata' '--enable-force-cgi-redirect' '--enable-pcntl' '--with-imap=shared' '--with-imap-ssl' '--enable-mbstring=shared' '--enable-mbregex' '--with-ncurses=shared' '--with-gd=shared' '--enable-bcmath=shared' '--enable-dba=shared' '--with-db4=/usr' '--with-xmlrpc=shared' '--with-ldap=shared' '--with-ldap-sasl' '--with-mysql=shared,/usr' '--with-mysqli=shared,/usr/bin/mysql_config' '--enable-dom=shared' '--with-pgsql=shared' '--enable-wddx=shared' '--with-snmp=shared,/usr' '--enable-soap=shared' '--with-xsl=shared,/usr' '--enable-xmlreader=shared' '--enable-xmlwriter=shared' '--with-curl=shared,/usr' '--with-curl-wrappers' '--enable-pdo=shared' '--with-sqlite=shared' '--enable-sqlite-utf8' '--with-pdo-odbc=shared,unixODBC,/usr' '--with-pdo-mysql=shared,/usr' '--with-pdo-pgsql=shared,/usr' '--with-pdo-sqlite=shared,/usr' '--enable-json=shared' '--enable-zip=shared' '--with-readline' '--enable-dbase=shared' '--with-pspell=shared' '--with-mcrypt=shared,/usr' '--with-mhash=shared,/usr' '--with-tidy=shared,/usr' '--enable-sysvmsg=shared' '--enable-sysvshm=shared' '--enable-sysvsem=shared' '--enable-posix=shared' '--with-unixODBC=shared,/usr' '--with-litespeed' '--enable-suhosin'

- upload dir is /tmp which is mounted on tmpfs
- suexec is on
- server is Plesk 9.5
- mysql is 5.1.62
- the issue will occur randomly but it will occur. So if you try, say 20 or 30 uploads, one of them will eventually fail.

 

[webizen]

... More specifically the asynchronous file upload script will not work (/wp-admin/async-upload.php) and the admin will get an

HTTP Error

error box.
...
Any help would be appreciated!

Any specifics regarding which line of async-upload.php it fails on?

 

[priestjim]

The script does not produce an error neither in the error_log nor in Litespeed's log. It just fails by return "HTTP Error".

 

[justme]

Hi,
any progress on this issue?
lsws standart 4.1.13 on ubuntu 11.10 64bit, wordpress 3.4.1
Client Ka M R In/Total Out/Total VHost Handler P Request
89.92.x 0 PR 222.6 119961/119961 0/0 x.fr php-fpm 223 POST /wp-admin/media-upload.php?inline=&upload-pag ..."
89.92.x 0 PR 201.4 120128/120128 0/0 x.fr php-fpm 202 POST /wp-admin/async-upload.php HTTP/1.1

uploadprogress.so has been added, php.ini mem is over 200MB allocated, ditto for max body size. Sometimes it works but most of the time it doesn't (result == stall). It got broken since 4.1.x (was OK in 4.0.x).
Thanks for your help!

 

[webizen]

make sure memory limit (soft/hard) in lsphp5 external app is big enough (larger than memory_limit in php.ini).

 

[justme]

make sure memory limit (soft/hard) in lsphp5 external app is big enough (larger than memory_limit in php.ini).

Actually, it is fastcgi/php-fpm, does it still apply?

 

[NiteWave]

if it's fastcgi/php-fpm, since I'm lack of the experience, may not help you much. the reason not use lsapi php ?

 

[justme]

if it's fastcgi/php-fpm, since I'm lack of the experience, may not help you much. the reason not use lsapi php ?

Because you don't have lsapi packages for Ubuntu? I did use lsapi with previous versions and went through the pain of compiling the package, but it is not practical for everyday's use with security fixes being pushed quite often. And since I am not concerned by performance issues and php-fpm is very close performance-wise, it would be nice that you support it since there's an obvious litespeed bug there and I don't feel like switching all conf to nginx/lighttpd

 

[webizen]

pls pm your server temp root ssh access if you like us to further troubleshoot.

 

[justme]

pls pm your server temp root ssh access if you like us to further troubleshoot.

I did, let me know when you are done