if you kill apache incorrectly it will NOT kill the fastcgi processes.
same for lighttpd.
httpd/dispatch are really not taking up that much memory. a lot of
that memory is shared, but there is no easy way to find out how much
of it.
as for the 2GB of swap being used im not exactly sure. maybe run top
to see what is using all that virtual space up. if its dispatch id
imagine your doing something (extremely greedy query perhaps) thats
causeing extensive memory usage.
On 8/30/05, hui <fortez-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:> 1.on my server,each "httpd" process is costing more than 15m mem
>
> 2.each "dispatch.fcgi" process is costing more than 20m mem
>
> 3.there is about 20 processes of ruby/dispatch.fcgi running at a time.
> and about 22 "httpd" processes.
> (using a command:ps aux |grep httpd |wc -l)
>
> 4.I believe there is something eating Mem,cos the server was down
> yesterday,the SWAP was full of 2G data
>
> 5.and I also notice,when I restarted httpd,the processes of
> dispatch.fcgi became 32.there is something has not released?
>
> I wannder how is your sever load?
> and how heavy are your "httpd" and "dispatch.fcgi"?
>
> many thanks^^
> _______________________________________________
> Rails mailing list
> Rails-1W37MKcQCpIf0INCOvqR/iCwEArCW2h5@public.gmane.org
> http://lists.rubyonrails.org/mailman/listinfo/rails
>
--
Zachery Hostens <zacheryph-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>