I'm trying to get a php installation configured with curl to work with Lite Speed.
I deleted my lsws installation.
Downloaded Version: LiteSpeed/2.0.3 Standard
Installed with PHP support.
Everything works fine.
I downloaded php 4.3.11 and made the change in cgi_main.c
Here is the code in mine after the change...
/* pre 4.3 behaviour, shouldn't be used but provides BC
if (env_path_info) {
SG(request_info).request_uri = env_path_info;
} else {
SG(request_info).request_uri = env_script_name;
}*/
SG(request_info).request_uri = env_script_name;
Here is my config, I took out gd, because I got tired of tracking configure and make errors
I'll take care of that later.
./configure --includedir=/usr/include --enable-fastcgi --with-config-file-path=../conf --disable-path-info-check --enable-discard-path --with-zlib --with-zlib-dir=/usr --enable-shmop --enable-track-vars --enable-sockets --enable-sysvsem --enable-sysvshm --enable-magic-quotes --with-curl
I took it directly from the phpinfo and took out gd and added curl.
It builds fine. I copied the original php in lsws/fcgi-bin/ to /lsws/fcgi-bin/php.org. I then copied the new cli to lsws/fcgi-bin/php.
I ran the following command from the fcgi-bin directory...
./php ../DEFAULT/html/phpinfo.php
which gave the expected output
I shutdown all php processes.
killall php
I then restarted the server from the admin interface and from the command line with and without turck installed. A clean installation for both.
kill -term <pid>
Now when accessing the default site's phpinfo from a browser it hangs and I get the following...
503 Service Unavailable
I get the proper 200 entry in my access log and no entries in the error log.
I have noticed that a 'ps aux | grep php' returns only the two admin_php entries and no others.
Running on a Redhat Enterprise 3 box.
Any Ideas?
On a side not, I love this webserver. I have set up our gaming clan on another box using it and the speed improvements over apache are nothing but amazing. Thanks.
I deleted my lsws installation.
Downloaded Version: LiteSpeed/2.0.3 Standard
Installed with PHP support.
Everything works fine.
I downloaded php 4.3.11 and made the change in cgi_main.c
Here is the code in mine after the change...
/* pre 4.3 behaviour, shouldn't be used but provides BC
if (env_path_info) {
SG(request_info).request_uri = env_path_info;
} else {
SG(request_info).request_uri = env_script_name;
}*/
SG(request_info).request_uri = env_script_name;
Here is my config, I took out gd, because I got tired of tracking configure and make errors
I'll take care of that later../configure --includedir=/usr/include --enable-fastcgi --with-config-file-path=../conf --disable-path-info-check --enable-discard-path --with-zlib --with-zlib-dir=/usr --enable-shmop --enable-track-vars --enable-sockets --enable-sysvsem --enable-sysvshm --enable-magic-quotes --with-curl
I took it directly from the phpinfo and took out gd and added curl.
It builds fine. I copied the original php in lsws/fcgi-bin/ to /lsws/fcgi-bin/php.org. I then copied the new cli to lsws/fcgi-bin/php.
I ran the following command from the fcgi-bin directory...
./php ../DEFAULT/html/phpinfo.php
which gave the expected output
I shutdown all php processes.
killall php
I then restarted the server from the admin interface and from the command line with and without turck installed. A clean installation for both.
kill -term <pid>
Now when accessing the default site's phpinfo from a browser it hangs and I get the following...
503 Service Unavailable
I get the proper 200 entry in my access log and no entries in the error log.
I have noticed that a 'ps aux | grep php' returns only the two admin_php entries and no others.
Running on a Redhat Enterprise 3 box.
Any Ideas?
On a side not, I love this webserver. I have set up our gaming clan on another box using it and the speed improvements over apache are nothing but amazing. Thanks.