From owner-freebsd-hackers Thu Nov 20 21:36:27 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id VAA14792 for hackers-outgoing; Thu, 20 Nov 1997 21:36:27 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id VAA14786 for ; Thu, 20 Nov 1997 21:36:23 -0800 (PST) (envelope-from jkh@time.cdrom.com) Received: from time.cdrom.com (jkh@localhost.cdrom.com [127.0.0.1]) by time.cdrom.com (8.8.7/8.6.9) with ESMTP id VAA01929; Thu, 20 Nov 1997 21:36:41 -0800 (PST) To: Jaye Mathisen cc: hackers@FreeBSD.ORG Subject: Re: Serious performance issue with 2.2.5-RELEASE In-reply-to: Your message of "Thu, 20 Nov 1997 12:17:36 PST." Date: Thu, 20 Nov 1997 21:36:41 -0800 Message-ID: <1925.880090601@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I upgraded my web server to 2.2.5 from 2.2.2-stable, dated sometime in > July. > > Big mistake. Perhaps, though if any clear mistakes were made here, it was in the transition process itself, it still remaining for me to be convinced that 2.2.5 itself is the culprit or I think I'd be hearing a lot more reports like this, especially from the 2.2.5 based web sites at some of the various ISPs I help maintain directly myself. The clear mistake was in hot-upgrading a production machine rather than installing/upgrading a 2nd box (or, at the very minimum, a second disk with / and /usr on it) and doing the cut-over as a staged exercise. Then if you'd seen the 2.2.5 problem here, you'd have simply switched back to the previous setup and given yourself the breathing room to puzzle over the new problem scenario at your leisure. Doing it the way you just did it is, frankly, about as completely and utterly wrong as it's possible to do things in an ISP environment and various protruding parts of you now need to be slapped, if you have someone around to provide this helpful service for you, so that you don't do something this criminally stupid in a production environment again. :-) As for your "weird pauses", that's the first I've heard of any such symptoms and would strongly recommend that you start trying to collect data during those pauses as to whether it's the interface, DNS, routes, what exactly "hangs" at the lowest level of this. Jordan