mod_python + pagetopdf (system call is not working)

[jc@…]

Inside pagetopdf, there is a system call to generate the PDF from html ussing htmldoc. That works find in my fcgi environment. After setting the Trac to mod_python, the system call is not evaluated.

(the temp files are created, but the result is empty!)

Any ideas? Do i need to set the PATH for a system call, or something similar?

Other modules and plugins work well. e.g. blog, imagegallery, webadmin etc, but another module (same as pagetopdf system) which spans a system call to generate formatted text using links/w3m or similar, is not working

Thanks in advance!

[Noah Kantrowitz]

Do you have SELinux or some other security system running, or is Apache running in a chroot?

[jc@…]

Nope.

Debian installation with apache 2.2 and mod_python. And apache is configured to run as nobody (as usual) but for cgi's its working. (even that process runs as nobody)

[jc@…]

It's me again, sorry for the second post.

Its working when i specify the complete path to the binary e.g. /usr/bin/htmldoc or similar. But the problem with the system call still exists, also with the enscript rendering.

[anonymous]

Woops. Permissions on EGG files and the paths where they end up ARE important. Closed.