Apache restart problem

[ircmaxell]

Something doesn't like Apache running on port 8080. I keep getting restart e-mails every 10 minutes saying that httpd failed. When I look at the apache error log, all it says is that apache starts, and detects that apache.pid has been overwritten. Is there any way I can get apache to play nice with 8080?

I need apache on 8080, because it's just too memory intensive as my front end server. I want to run lighttpd on port 80 to handle all files but .cgi, .php, .html (I am using mod_rewrite in apache to make my php files look like .html files). I then use mod_proxy to pass the request back to apache if it falls into those three. The site gets a very large amount of request for xml and htm files durring certain times (about 50 requests/second). I can't get apache's children down under 100 megs (after it runs for a few minutes)...

[xerophyte]

My guess is chkservd, which monitor the apache and restart. You can configure to not to monitor using whm i guess.

hope that helps

[ircmaxell]

Well, I like the monitor... Would there be any way to make it check port 8080 instead of 80? And I would prefer to integrate lighttpd into chkservd as well...

[ircmaxell]

Alright, I've got it fixed... Sorta... I changed /etc/chkserv.d/httpd to watch on 8080 (works fine)... Then I added "lighttpd:1" to /etc/chkservd/chkservd.conf and this to a file lighttpd

service[lighttpd]=80,GET / HTTP/1.0,HTTP/1..,killall -TERM lighttpd;killall -9 lighttpd;/usr/local/sbin/lighttpd -f /etc/lighttpd/lighttpd.conf

It is now killing lighttpd every 8 minutes and restarting it... That entire line (killall -TERM... etc) works from shell... My question is as follows
1. Why is it killing lighttpd? I've checked the response and it's HTTP/1.0 403 Forbidden, so that should still work with http/1.. right?