Received: by taz.hyperreal.com (8.6.12/8.6.5) id JAA06521; Fri, 8 Dec 1995 09:15:43 -0800 Received: from ooo.lanl.gov by taz.hyperreal.com (8.6.12/8.6.5) with ESMTP id JAA06514; Fri, 8 Dec 1995 09:15:40 -0800 Received: by ooo.lanl.gov (1.37.109.16/16.2) id AA106082927; Fri, 8 Dec 1995 10:15:27 -0700 From: Rob Hartill Message-Id: <199512081715.AA106082927@ooo.lanl.gov> Subject: Re: patch 20 death To: new-httpd@hyperreal.com Date: Fri, 8 Dec 95 10:15:27 MST In-Reply-To: ; from "David Robinson" at Dec 8, 95 4:59 pm X-Organization: Theoretical Division, T-8. Los Alamos National Laboratory X-Snail: LANL Theoretical Divi' T-8, MS B285, P.O Box 1663, Los Alamos NM 87545 X-Marks-The-Spot: Doh ! X-Url: http://nqcd.lanl.gov/~hartill/ X-Cessive-Use-Of-Headers: check Mailer: Elm [revision: 70.85] Sender: owner-new-httpd@apache.org Precedence: bulk Reply-To: new-httpd@apache.org > I've been out of touch for a week. As this is urgent, can someone summarise > the observed problems please? (Probably quicker than me wading through the > ~400 messages in my inbox). ok here goes... Apache spawns to death in most cases, possibly filling the error_log too. My personal observations (possibly unrelated, but who knows..) My sunos machine at Cardiff started comitting suicide when we moved to 0.8.15. In this case the error_log filled with bogus messages about access denied, which were recorded as coming from the same client address. the error_log filled and took out the server which didn't have any diskspace left to write to (we do lots of disk writing CGI). The access denied was on html files that used SSI. This happened a couple more times before I added a 1 minute error_log crontab to catch a runaway error_log and sighup Apache. this cron was catching one of these every day or three. On one occasion, after a machine reboot, the machine melted due to lack of memory. The machine at Cardiff had MaxClients set to 18, and still the problem continued. When I recompiled 1.0.0 without mod_user and mod_negotiation the problem disappeared. Someone suggested that the symptoms could relate to the scoreboard getting nuked, so we found patch 20 to be a candidate. People using Sunos, linux, *BSD* (I can't keep up with these names), who experienced the problem have reversed patch 20. So far, there are no reports that the problem exists after 20 is removed. We're sure the problem came in after 0.8.14. rob