You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Rob Hartill <ha...@hyperreal.com> on 1995/08/29 19:16:07 UTC

Linux, linux everywhere....

Forwarded message:
> From joe@volant.com  Mon Aug 28 23:30:34 1995
> Message-Id: <19...@volant.com>
> From: Joe Austin <jo...@volant.com>
> Date: Mon, 28 Aug 95 23:20:08 -700
> To: apache-bugs@apache.org
> Mime-Version: 1.0
> X-Mailer: Mozilla/1.0N (Windows)
> Content-Type: text/plain;  charset=iso-8859-1
> Content-Transfer-Encoding: 8bit
> Subject: Server Stops Responding And Recovers
> 
> 
> Greetings.
> 
> We have been using Apache on all of our web servers
> (currently 8 Pentium systems running Linux).  Collectively,
> these servers get over 1 million+ hits per day.  They
> all share a common Linux/Pentium NFS dedicated disk server.
> 
> On the heavily traveled servers, we experience Apache
> not responding for several minutes and then responding
> again.
> 
> This *is not* a case of too many processes or system
> load problems.  There are (usually) no more than
> 10-20, or so, processes running when the problem occurs.
> 
> Other TCP/IP services work fine while the problem occurs.
> 
> If anyone on the Apache Team would like to see the problem
> first hand, then we would gladly open up our development
> and testing environment.
> 
> Best Regards,
> 
> Volant Corporation
> Joe Austin
> joe@volant.com
> 
> 


Re: Linux, linux everywhere....

Posted by Brian Tao <ta...@gate.sinica.edu.tw>.
On Tue, 29 Aug 1995, Rob Hartill wrote:
> 
> > On the heavily traveled servers, we experience Apache
> > not responding for several minutes and then responding
> > again.

    Sounds like exactly the same problem the folks at
ftp.cica.indiana.edu are (or were?) experiencing on their Linux-based
server with wu-ftpd.  The server would just hang for several minutes
on end, even while a second copy of ftpd continued to run happily on a
different port.

    I know this doesn't help directly, but perhaps the CICA people
have found a solution to this long-standing Linux problem (increasing
the listening queue length, or patch xinetd in ftpd's case, or switch
to FreeBSD ;-) ).
-- 
Brian ("Though this be madness, yet there is method in't") Tao
taob@gate.sinica.edu.tw <-- work ........ play --> taob@io.org