httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ben Laurie <...@gonzo.ben.algroup.co.uk>
Subject Re: Apache 1.2b7-dev performance
Date Sat, 08 Feb 1997 23:49:18 GMT
Marc Slemko wrote:
> 
> Hang on a minute here.  lingering_close() is NOT just intended to fix a
> problem with PUTs.  As I have said before, there are other workarounds
> which can be done to solve the PUT problem without lingering_close() and
> there are other workarounds to solve the SVR4 problem where sockets are
> destroyed on close() before the data is sent.  If those were the only
> problems, we could get rid of lingering_close().
> 
> The problem is persistent connections.  There are times when the server
> closes the connection.  The main ones are:
> 
> 	- keepalive timeout
> 	- "unusual" response that results in the server closing the
> 	  connection.  Currently for Apache those include things like
> 	  server errors, redirects. etc.
> 	- max number of requests per connection; used to be a big
> 	  issue, but not anymore unless someone sets that low.
> 
> Whenever the server closes the connection, there is a risk of the 
> client getting an error if it tries to send data before it knows the
> connection is closed.

But the question is, why do we care if the client gets an error? After all, we
got an error already...

>  I have been through this half a dozen time before
> but have got neither any reason why my statement is wrong nor anyone
> who says "oh, that makes sense".  I am trying to figure out how the heck
> I can say it to make people understand, but am having some trouble without
> anyone saying exactly what they don't understand.  Roy also explains
> some of it on the fin_wait_2.html page.
> 
> Note that this problem _may_ (ie. I'm not sure yet) only occur
> during out of order packet delivery.
> 
> I'm in the process of making a test client to try to reproduce
> the problem.
> 
> On Sat, 8 Feb 1997, Randy Terbush wrote:
> 
> > 
> > I'm beginning to wonder if we would not be better served to
> > make NO_LINGCLOSE the default behavior and allow lingering_close()
> > as an option for sites that are effected by the PUT problem it was
> > intended to fix. My feeling is that the PUT problem is far less
> > common than the need for speed.
> > 
> > 
> > > I tried compiling with the NO_LINGCLOSE option and sure enough, Apache
> > > 1.2b7-dev (with Deans' patches + my mod_include patch) performed as well as
> > > Apache 1.1.3.
> > > 
> > > I'd like to add a section to the documentation on improving performance,
> > > describing lingering_close, why we've added it, and the possibility that using
> > > NO_LINGCLOSE will substantially increase web server performance.  Under certain
> > > rather specific conditions, it appears that the increase in capacity might
be
> > > as high as 50%; a more normal improvement is around 10-20%.
> > > 
> > > -- 
> > >      -- Ed Korthof        |  Web Server Engineer --
> > >      -- ed@organic.com    |  Organic Online, Inc --
> > >      -- (415) 278-5676    |  Fax: (415) 284-6891 --
> > 
> > 
> > 
> 

-- 
Ben Laurie                Phone: +44 (181) 994 6435  Email: ben@algroup.co.uk
Freelance Consultant and  Fax:   +44 (181) 994 6472
Technical Director        URL: http://www.algroup.co.uk/Apache-SSL
A.L. Digital Ltd,         Apache Group member (http://www.apache.org)
London, England.          Apache-SSL author

Mime
View raw message