Rails - Aug 2006 - Mongrel 0.3.13.4 Pre-Release -- PID Files Working

Hi Folks,

An update to tell everyone to try out the pre-release again:


  sudo gem install mongrel
--source=http://mongrel.rubyforge.org/releases/


Nothing available for win32 yet, the next pre-release will have a win32
build to check that windows folks are OK.

This release fixes the following problems:

* PID files weren''t being written when configured.  Works now and logs
messages better.
* Small change to the HTTP parser to allow for headers that don''t have
a
space after the ":" character, which is allowed by the RFC.
  -- Everyone using McAfee Security Center should contact them and tell
them their product is a piece of crap.  You should also consider not
using it since they seem to be modifying people''s HTTP requests in
bogus
ways and telling you that this is a "security" feature.

This release is pretty close to becoming official.  Everyone needs to
test it out and shoot me any final problems.  The current outstanding
problems I have are:

 55.BUG: test tom brice''s upload app with keepalive
 56.BUG: safari ganks on the deflate filter
 57.BUG: look at Ben Moore''s bug on ECONNABORTED
 59.BUG: content-type not getting set

If the folks mentioned could check this release against their problems
then that will help quite a bit.

Thanks!


-- 
Zed A. Shaw
http://www.zedshaw.com/
http://mongrel.rubyforge.org/
http://www.railsmachine.com/ -- Need Mongrel support?

On 8/11/06 1:58 PM, Zed Shaw  wrote:
> This release is pretty close to becoming official.  Everyone needs to
> test it out and shoot me any final problems.  The current outstanding
> problems I have are:
> 
>  55.BUG: test tom brice''s upload app with keepalive
Zed,
I tested the app I sent you and it appears to be fixed.  I''m having a
few
"lags" here and there but the 4 mongrel''s I''m
running do not appear to lock
or run forever.  Have commented out "KeepAlive Off" in my virtual host
section and my browser confirms "Keep-Alive timeout=5, max=100".  I
would
like to figure out how to tweak those values but that is a question for the
apache folks.

My testing is by no means scientific but it seems to be resolved.

Thanks,
Tom