From owner-freebsd-net Tue Oct 30 11: 4:29 2001 Delivered-To: freebsd-net@freebsd.org Received: from silby.com (cb34181-a.mdsn1.wi.home.com [24.14.173.39]) by hub.freebsd.org (Postfix) with ESMTP id 7D1AF37B401 for ; Tue, 30 Oct 2001 11:04:27 -0800 (PST) Received: (qmail 5657 invoked by uid 1000); 30 Oct 2001 19:04:22 -0000 Received: from localhost (sendmail-bs@127.0.0.1) by localhost with SMTP; 30 Oct 2001 19:04:22 -0000 Date: Tue, 30 Oct 2001 13:04:22 -0600 (CST) From: Mike Silbersack To: veedee Cc: "freebsd-net@FreeBSD.ORG" Subject: Re: your mail In-Reply-To: <20011030184347.7EBE537B401@hub.freebsd.org> Message-ID: <20011030125624.E5542-100000@achilles.silby.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Tue, 30 Oct 2001, veedee wrote: > [#] netstat -m > 309/1232/6144 mbufs in use (current/peak/max): > 194 mbufs allocated to data > 115 mbufs allocated to packet headers > 170/352/1536 mbuf clusters in use (current/peak/max) > 1012 Kbytes allocated to network (21% of mb_map in use) > ... > > what would be an appropriate value? I'm running FreeBSD 4.3 on this box and I have about 400 workstations on my "neck". > > Thanks in advance, > Radu Bogdan Rusu (aka veedee) > C7 Campus Network System Administrator There's no exact science, but the tuning manpage seems to give a pretty good formula to calculate what you could use. It'd be easier to just pick an abritrary number like 6000, though. :) And if the problem does happen again, run a netstat -n to see if you can see a pattern to buffer usage; there is a DoS called netkill which can be used to suck up all network buffers and cause the problem you're seeing. If it's being used, you'll see that one IP is eating up all the buffers. Mike "Silby" Silbersack To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message