From Bugzilla Helper: User-Agent: Mozilla/5.0 Galeon/1.2.5 (X11; Linux i686; U;) Gecko/20020623 Debian/1.2.5-0.woody.1 Description of problem: When a variables is passed on PHP on Apache 2.0.40, the variable's value is not set correctly. It may be a php bug, but I have been told that it just might be an apache bug. To verify a script in perl can be written to use POST method similar to the one on the URL posted. Version-Release number of selected component (if applicable): httpd-2.0.40-21.3 How reproducible: Always Steps to Reproduce: 1a. Install RH 8, It's a bug here too, if I remember correctly 1b. Install RH 8 2b. Use apt for rpms to upgrade to RH 9 Bug should occur still 1c. Install RH 9 Bug should occur 1d. Instal RH 8 2d. upgrade to RH 9 Bug should occur 1e. Install RH 8 2e. Upgrade to RH 9 through apt for rpms Bug should still occur 1f. Install RH 8 2f. Upgrade to RH9 through apt for rpms 3f. Use up2date to upgrade system Bug should still occur Actual Results: I get as output for the license variable: "I acceptlicense=I accept" Expected Results: Correct output should be "I accept" Additional info: apache stable 2.0.46 is out. RH does not yet have that rpm available. I've asked on irc.redhat.com about the bug on #apache and #php and no one was sure of the bug. PHP people blamed apache. I'm noting it as with high priority since it ruins many POST method web apps and there is no easy way out of it.
The okuili.com server on which the test page was on is now down and that is because I am installing Debian on it now. But, the test page/tarball is still available at: http://ruslug.rutgers.edu/~mcgrof/test/reg.php
Have you added "AddType application/x-httpd-php .php" to your httpd.conf? You shouldn't add that for the PHP package in Red Hat Linux 8.0 and 9. This is probably a duplicate of bug 82967.
If I remember correctly, I did have that option set. If the other bug had that as the main problem, then yes this is probably, 80% sure, a duplicate bug. Unfortunately, this was not obvious to the end user. I can't imagine the amount of RH machines out there with the same setup.
Yes, a forthcoming erratum will allow PHP to work correctly even if that AddType directive is configured. I'll close this as a duplicate; thanks for submitting the report. *** This bug has been marked as a duplicate of 82967 ***
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.