From owner-freebsd-stable Wed May 5 6: 3:30 1999 Delivered-To: freebsd-stable@freebsd.org Received: from europe.std.com (europe.std.com [199.172.62.20]) by hub.freebsd.org (Postfix) with ESMTP id 7BD2115027 for ; Wed, 5 May 1999 06:03:27 -0700 (PDT) (envelope-from lowell@world.std.com) Received: from world.std.com by europe.std.com (STD1.2/BZS-8-1.0) id JAA06379; Wed, 5 May 1999 09:03:13 -0400 (EDT) Received: by world.std.com (TheWorld/Spike-2.0) id AA08128; Wed, 5 May 1999 09:03:13 -0400 To: freebsd-stable@freebsd.org Cc: Greg Lehey , Doug White Subject: Re: Current state of the conventional wisdom? References: <19990504092501.G10134@freebie.lemis.com> <19990505110655.U40359@freebie.lemis.com> From: Lowell Gilbert Date: 05 May 1999 09:03:13 -0400 In-Reply-To: Greg Lehey's message of Wed, 5 May 1999 11:06:55 +0930 Message-Id: Lines: 46 X-Mailer: Gnus v5.5/Emacs 20.2 Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG [redirected to -stable rather than -questions] Greg Lehey writes: > On Tuesday, 4 May 1999 at 15:17:34 -0700, Doug White wrote: > > 1. 2.2.X -> 3.X upgrade doesn't handle a.out libraries, rendering a.out > > bins (netscape, built ports, etc.) useless until they're moved. > > It did for me. I am pretty sure that this wasn't done until some time in January or February. Before the release of 3.1, though, for sure. > > > 2. Sysinstall locks up during the second device probe. > > I haven't seen this. > > > 4. Kernel panics mysteriously after device probe when devices are > > removed; unedited kernel boots normally. > > I haven't seen this. I've seen both, and I've been trying to track them down. Unfortunately, I see them on my very old home machine, and nowhere else. That machine has very little CPU, and very slow disks, so experiments take a very long time (as a benchmark, note that a minimal CD install from scratch would take an hour, even if it went smoothly). The fact that I get complete lockups or spontaneous reboots a lot more than I get kernel panics also slows down debugging. When the system's up and running, tasks will tend to die with bus errors (signal 10), so I don't have full confidence in the hardware either (despite the fact that it's still rock-solid under 2.2.2). I hadn't actually been sure that disabling devices by configuration was connected to failing to boot the kernel, but I'd kind of had that impression. The heisenbug nature of the failures I'm seeing makes it hard to establish cause and effect. Given that I'm sure I could make all these problems go away for a couple of hundred dollars or so in new hardware, there's a limit to how much time I'm willing to spend watching the screen display boot messages. Be well. Lowell To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message