Rails - Jun 2006 - Dispatch.fcgi h-e-double hockey sticks...

Hello Everyone ~

Last week I started having a dispatch.fcgi process go out of control on my
Linux server.  If I don''t catch it the process will eat my RAM, and
devour
my swap file until the machine becomes non-responsive.  I am running
Lighttpd and have not had any issues until lately.

The logs are not telling me anything useful, and as a quick fix I am running
the site using Mongrel, which I am very impressed with!  I would however
like to use Light, as I also have some PHP sites on the box as well.

Any help or pointers on how to debug this issue are greatly appreciated!

Thx,

Ben

-- 
Ben Reubenstein
303-947-0446
http://www.benr75.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://wrath.rubyonrails.org/pipermail/rails/attachments/20060626/46db323e/attachment.html

On Mon, 2006-06-26 at 16:37 -0600, Ben Reubenstein
wrote:
> Hello Everyone ~ 
> 
> The logs are not telling me anything useful, and as a quick fix I am
> running the site using Mongrel, which I am very impressed with!  I
> would however like to use Light, as I also have some PHP sites on the
> box as well.  
Best bet in this case is to continue running mongrel and lighty, but put
pen or balance between them so that you get around lighty''s broken
mod_proxy balancing.  It''d look something like this:


 {interwebs} -->  [lighttpd] -> [balance] --> [mongrel1, mongrel2,
mongrel3, ...]

There were several folks who said they did that or planned to do that
(DHH himself), but I didn''t get any definitive results in their
stories.

Try it out and see how it works.  Also use mongrel_cluster to manage the
mongrel servers.

-- 
Zed A. Shaw
http://www.zedshaw.com/
http://mongrel.rubyforge.org/