Return-Path: Delivered-To: apmail-httpd-users-archive@httpd.apache.org Received: (qmail 67391 invoked by uid 500); 28 Jul 2003 18:33:29 -0000 Mailing-List: contact users-help@httpd.apache.org; run by ezmlm Precedence: bulk Reply-To: users@httpd.apache.org list-help: list-unsubscribe: list-post: Delivered-To: mailing list users@httpd.apache.org Received: (qmail 67377 invoked from network); 28 Jul 2003 18:33:29 -0000 Received: from out2.smtp.messagingengine.com (HELO mail.messagingengine.com) (66.111.4.26) by daedalus.apache.org with SMTP; 28 Jul 2003 18:33:29 -0000 Received: from mail.messagingengine.com (localhost [127.0.0.1]) by localhost.localdomain (Postfix) with ESMTP id 4A64F69D89 for ; Mon, 28 Jul 2003 14:33:33 -0400 (EDT) Received: from 10.202.2.150 ([10.202.2.150] helo=mail.messagingengine.com) by messagingengine.com with SMTP; Mon, 28 Jul 2003 14:33:33 -0400 X-Epoch: 1059417213 X-Sasl-enc: KSud2hYLQy+bbQKiTWA1Og Received: from usager70-65.hec.ca (usager70-65.hec.ca [132.211.70.65]) by www.fastmail.fm (Postfix) with ESMTP id 0391969C1A for ; Mon, 28 Jul 2003 14:33:33 -0400 (EDT) Date: Mon, 28 Jul 2003 14:31:18 -0400 (=?ISO-8859-1?Q?Est_=28heure_d'=E9t=E9=29?=) From: Joshua Slive To: users@httpd.apache.org In-Reply-To: <20030728151938.5518d500.msarsale@buenosaires.gov.ar> Message-ID: References: <20030728151938.5518d500.msarsale@buenosaires.gov.ar> X-X-Sender: slive@www.fastmail.fm MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N Subject: Re: [users@httpd] machine slowdown On Mon, 28 Jul 2003, Martin Sarsale wrote: > This is the problem: we're having a big number of connections (about 100 > simultaneous clients with peaks of 250 (then, MaxChildren is reached)) > and after the server is running for some time, it stops answering > requests and the system becomes _very_ slow: it's not possible to log in > via ssh and when login via the console everything works really slow > (some seconds pass between each program executed). > Mem: 255516K av, 252064K used, 3452K free, 0K shrd, 1548K buff > Swap: 408796K av, 350988K used, 57808K free 14328K cached You are out of memory. The system is swapping back and forth to disk and thrashing itself to death. To verify this, there is probably a program on linux that will tell you the exact swapping activity. I don't remember what it is, however. Some ways you can fix the problem: 1. Reduce MaxClients. You said you did this, but obviously you haven't reduced it enough since you are still out of memory. 2. Buy more memory. 3. Try to trim down apache so that it takes less memory. Do you really need all those modules? How much memory is each httpd using? 4. Try apache 2 with a threaded mpm like worker (but make sure all the modules you need are supported). 5. You can also try reducing the KeepaliveTimeout, which should help get rid of some of the requests in the "K" state in your server status report. But that probably won't be enough to solve your problem, since most of the requests are in "W" (sending response). Joshua. --------------------------------------------------------------------- The official User-To-User support forum of the Apache HTTP Server Project. See for more info. To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org " from the digest: users-digest-unsubscribe@httpd.apache.org For additional commands, e-mail: users-help@httpd.apache.org