From owner-freebsd-hackers Mon Feb 5 14:22:40 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by hub.freebsd.org (Postfix) with ESMTP id 349F337B401; Mon, 5 Feb 2001 14:22:21 -0800 (PST) Received: (from bright@localhost) by fw.wintelcom.net (8.10.0/8.10.0) id f15MMIj13403; Mon, 5 Feb 2001 14:22:18 -0800 (PST) Date: Mon, 5 Feb 2001 14:22:18 -0800 From: Alfred Perlstein To: Dan Phoenix Cc: scanner@jurai.net, freebsd-hackers@FreeBSD.ORG, freebsd-question@FreeBSD.ORG Subject: Re: qmail IO problems Message-ID: <20010205142218.I26076@fw.wintelcom.net> References: <20010205135501.H26076@fw.wintelcom.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: ; from dphoenix@bravenet.com on Mon, Feb 05, 2001 at 02:05:01PM -0800 Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG * Dan Phoenix [010205 14:05] wrote: > > > Just tried both parameters....nada. > I am desperately trying to work out a solution here...because this machine > will be forced back to linux of i can;t find a way to improve the disk > I/O. Checking on qmail queue patches maybe? Far as I know there is a > big-to do patch on qmail's homesite....not sure if that will help but > worth a shot...definately not looking forward to seeing linux again! What happens when you run the fs in completely async mode? What is the output of your 'mount' command? How long did you run with those sysctl options? Your reply came very quickly, so I'm assuming not very long. Are you absolutely sure there's no configuration difference between when it was Linux and now that it's FreeBSD? Hardware _and_ software config (qmail) wise? Is btw, is this mail being queued up to be sent? Or is it mail being queued up for sending? Are you sure nothing went crazy during the changeover such that the load is now a lot higher? Such as: Are you sure your backup MX is not killing you trying to deliver all the queued mail it got while the main MX was down? Also, the disk isn't at 80-95%, those are transactions per-second. So unless you're actually seeing a performance problem, then you don't need to worry although that many tps is pretty high. Also, please don't post to -hackers and -questions. A question should be sent to either, not both. -- -Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] "I have the heart of a child; I keep it in a jar on my desk." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message