I don't find any threads where this specific question was previously raised. Is there some particular reason why a URL must be the argument of the lynx command? I have amember installed in a subdomain that requires SSL. The cron job, as specified in the amember doc, did not work. So I then tried the shell command in a terminal. Lynx returned error messages that indicated it could not run the file. I then found an old thread in the forum that indicates lynx does not support SSL, which would seem to explain the problem. I then tried running the lynx command using a directory path to cron.php instead of using the URL. This time lynx ran and dumped the script source on the terminal, as requested by the -source switch. Is there any reason running this way will not obtain the correct results?
Lynx is a text browser for the World Wide Web so it should accept url as argument (and not file path) What error message you get exactly when tryed to execute command from shell?
I found a simple workaround to the problem. Possibly it may help someone else. I simply put the following code in the main site VirtualHost configuration: Options +FollowSymLinks RewriteEngine On # Rewrite for cron RewriteCond %{REQUEST_URI} ^/obscure_name/$ RewriteRule ^(.*)$ /path/to/amember/cron.php [L] <Directory /path/to/amember> Order deny,allow Allow from all </Directory> Recall that the subdomain with amember is TLS-only. However, rewriting "http://www.mydomain.com/obscure_name/" to the cron script path avoids TLS entirely. Works a treat.