From owner-freebsd-hackers Mon Nov 13 06:25:44 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id GAA25739 for hackers-outgoing; Mon, 13 Nov 1995 06:25:44 -0800 Received: from ns0.netcraft.co.uk (ns0.netcraft.co.uk [194.72.238.4]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id GAA25719 ; Mon, 13 Nov 1995 06:25:38 -0800 Received: (from paul@localhost) by ns0.netcraft.co.uk (8.6.12/8.6.9) id OAA08697; Mon, 13 Nov 1995 14:19:37 GMT From: Paul Richards Message-Id: <199511131419.OAA08697@ns0.netcraft.co.uk> Subject: Re: FreeBSD 2.0.5R hanging inexplicably... To: paul@trumpet.net.au (Paul Reece) Date: Mon, 13 Nov 1995 14:19:37 +0000 (GMT) Cc: scrappy@hub.org, freebsd-questions@FreeBSD.org, freebsd-current@FreeBSD.org, freebsd-hackers@FreeBSD.org In-Reply-To: from "Paul Reece" at Nov 13, 95 03:03:46 pm Reply-to: paul@netcraft.co.uk X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 2991 Sender: owner-hackers@FreeBSD.org Precedence: bulk In reply to Paul Reece who said > > On Sun, 12 Nov 1995, Marc G. Fournier wrote: > > > For the second time in two days, my system just hung, in that, > > essentially, no keyboard input was being permitted. > > Which version are you running? I had this happen a lot with 2.0.5 since it didn't handle keyboard disconnection and unplugging and putting the keyboard back made it dissappear. I also had it happen "randomly" quite frequently but I put that down to the some quirk triggering the same problem. When I upgraded to stable the problem went away BUT I did have it happen to me once about three days ago shortly after reboot (5 mins say) just after upgrading to stable again. That's the only time it's happened in the last 6 weeks or so but there may be a problem lurking in there somewhere. This is with PCVT. I've had worse problems with syscons. There's something wrong with syscons vt switching, it's not completely reproducable but it happens frequently enough that I've now switched all our machines to pcvt. The symptom is that when you switch VT the screen goes completely blank and further attempts to switch VT just gets a beep (as though no VT was configured). Everything is still running (getty etc) but the screen is just totally blank. Only way to recover the console is to reboot. > > The first time, I'm sure it had to do with PCVT, in that it happened > > the moment after I had modified /etc/ttys and turned off all the console > > screens that I'm not using and issued a kill -1 1. Even after reboot, it > > hung almost immediately. When I booted up into single user mode, and > > turned those ports back on, it was all fine. > > > > Of significance to note in that case, everything ran fine in the > > background, it was just a console lock. The keyboard lock problems I've seen have been purely keyboard problems. I can start X terms on other machines and have them appear on the locked console so the display is still working fine. > > That is the symptoms...now the question...are there any known > > problems with PCVT under 2.0.5R that have been fixed in 2.1.0-*-SNAP? > > If not...how do I debug something like this? Since it hangs, and doesn't > > fault itself, getting a core dump, so I'd think, is kind of difficult... I'd say yes, I had this keyboard problem happen regularly with 2.0.5 and apart from this one lockup I had a few days ago everythings been fine since upgrading to 2.1-stable. I doubt the lockup I had a few days ago is the same problem since the old problem would happen every few hours, for the moment I'm putting this last lockup down to the phase of the moon :-) A nice feature would be an ioctl that caused the console driver to reset itself so that you can log in across the net and run some user-land util to reset the console driver without having to reboot. -- Paul Richards, Netcraft Ltd. Internet: paul@netcraft.co.uk, http://www.netcraft.co.uk Phone: 0370 462071 (Mobile), +44 1225 447500 (work)