httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Cliff Woolley <cliffwool...@yahoo.com>
Subject Re: tarballs are up
Date Tue, 31 Jul 2001 04:23:07 GMT
On Mon, 30 Jul 2001, Greg Ames wrote:

> Could you do me a huge favor?  attach to the parent with gdb, set a
> breakpoint in perform_idle_server_maintenance, and see what you have for
> idle_thread_count at this line:
>
>    if (idle_thread_count > max_spare_threads) {


Okay, here's what I've got.  While doing this:

jcw5q@cobra:/uf2/jcw5q/apache/test/bin$ ./ab -n 50000 -c 50
http://cobra.cs.virginia.edu:3333/manual/index.html.en
This is ApacheBench, Version 1.3e <$Revision: 1.80 $> apache-2.0
Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd,
http://www.zeustech.net/
Copyright (c) 1998-2001 The Apache Software Foundation,
http://www.apache.org/

Benchmarking cobra.cs.virginia.edu (be patient)
Completed 5000 requests
Completed 10000 requests
Completed 15000 requests
Completed 20000 requests
Completed 25000 requests
Completed 30000 requests


I did two of these back-to-back without letting the first one finish:

jcw5q@cobra:/uf2/jcw5q$ apachectl graceful
/uf2/jcw5q/bin/apachectl graceful: httpd gracefully restarted
jcw5q@cobra:/uf2/jcw5q$ apachectl graceful
/uf2/jcw5q/bin/apachectl graceful: httpd gracefully restarted


And got this:

______________W___________________________________..............
__________________________________________________..............
____C_____________________________________________..............
...........................G....................................
................................................................
................................................................
................................................................
................................................................

Breakpoint 2, perform_idle_server_maintenance () at threaded.c:1034
1034        if (idle_thread_count > max_spare_threads) {
(gdb) p idle_thread_count
$9 = 144
(gdb) p max_spare_threads
$10 = 55

I *think* the server status output I've given here pretty much lines up in
time with the breakpoint, but I can't be 100% sure because that breakpoint
got triggered so frequently that I was having trouble doing all of the
right things in the right sequence to trigger the bug.

PS: After continuing from the breakpoing, my browser had long since given
up its keepalive... and the server seems to have recovered immediately
afterward.  Okay, it definitely recovered, I'm just assuming it's the
dropping of the kept-alive connection that triggered the recovery.  Let me
know if I can try anything else.

Thanks,
--Cliff


--------------------------------------------------------------
   Cliff Woolley
   cliffwoolley@yahoo.com
   Charlottesville, VA



Mime
View raw message