httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Hartill <r...@imdb.com>
Subject WWW Form Bug Report: "Creation/Hanging of second 'root' owned httpd" on OSF/1 (fwd)
Date Tue, 10 Dec 1996 18:18:03 GMT
----- Forwarded message from develop@vaxxine.com -----

Message-Id: <199612101758.JAA27172@taz.hyperreal.com>
From: develop@vaxxine.com
To: apache-bugs%apache.org@organic.com
Date: Tue Dec 10  9:58:12 1996
Subject: WWW Form Bug Report: "Creation/Hanging of second 'root' owned httpd" on OSF/1

Submitter: develop@vaxxine.com
Operating system: OSF/1, version: 2.0
Version of Apache Used: 1.2b2
Extra Modules used: 
URL exhibiting problem: http://www.vaxxine.com, http://home.vaxxine.com (slow link on home)

Symptoms:
--
Yes, it's me again. :)  The problem here occured
in Apache 1.1 and continues into the 1.2 betas.
If someone could point me to the functions that
may cause an oddity like this I could probably
attempt to debug it.  The problem:  From time to
time a second "root" owned httpd seems to 
appear out of nowhere.  It runs out of control
consuming about 80% of the CPU.  Apache
continues to answer requests, albeit slowly.  If
I kill the out of control httpd Apache no longer
answers requests.  The "not out of control" 
root httpd continues as do the copies running
as "nobody" and something answers
on the port but no documents are returned.  
I know we are not attempting to start Apache
twice (it couldn't bind to the port anyway) so
there seems to be a fork() that can occur
before the switch of user IDs.  The problem
only seems to occur under heavy load.  
Please also note that if any of the developers
specialize in OSF/1 I would be happy to provide
him/her with a "telnet" account on one of our
quieter servers for any diagnostics that s/he
may wish to do.
--

Backtrace:
--

--

----- End of forwarded message from develop@vaxxine.com -----

-- 
Rob Hartill.       Internet Movie Database Ltd.    http://www.imdb.com/  

Mime
View raw message