I
have a feeling that I’m missing something elementary here. I have
an install of apache 2.0.55 with mod_ssl enabled on a HP-UX system in
/opt/apache2. This one runs fine. I recompiled another copy
of apache (same version) into /opt/apache2a (for testing purposes) to add
mod_ldap support and that one worked as well. Then I tried recreating
apache2a in apache2 by doing a recompile using a prefix of apache2 and then doing
an install after backing everything up and moving the old apache install out of
the way. However, this one DOESN’T work. If I launch it
WITHOUT SSL turned on (i.e, no SSLEngine on) directive, everything works great.
But as soon as I turn on SSL in a VirtualHost, then strange things
happen. A client will connect to the test port via SSL, the SSL negotiation
appears to work just fine (tested using openssl s_client), but when you attempt
to do a GET, the request is sent, but a reply never shows up. Nothing
appears in the access_log, and child processes begin to spawn with each
request. I can pull up the server-status url and everytime I hit refresh,
one child process goes to “W” and another one is spawned. Clicking
repeatedly will continue this process until there are a ton of processes, all
stuck at “Waiting” with 0/0/0 under the Acc columntDo it enough,
the server’s load average starts to climb. I’ve
checked and double checked every permission I can possible find. The User
and Group directives are both set to “webadmin” which is the same
in all configurations. The permissions of the sub-directories in both
directories match between the two. I have this feeling that it’s
simple with the directory permissions and/or structure but I just can’t
seem to locate it. Anyone have any ideas on what else I might need to
look at? Aaron |