From owner-freebsd-questions Fri Nov 10 13: 7:23 2000 Delivered-To: freebsd-questions@freebsd.org Received: from guru.mired.org (okc-65-26-235-186.mmcable.com [65.26.235.186]) by hub.freebsd.org (Postfix) with SMTP id 616B637B479 for ; Fri, 10 Nov 2000 13:07:19 -0800 (PST) Received: (qmail 46097 invoked by uid 100); 10 Nov 2000 21:07:18 -0000 From: Mike Meyer MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14860.25478.379016.781155@guru.mired.org> Date: Fri, 10 Nov 2000 15:07:18 -0600 (CST) To: Stanley Hopcroft Cc: questions@freebsd.org Subject: Re: 4.1-RELEASE. panic. I'm not going anywhere without my init. In-Reply-To: <129162557@toto.iv> X-Mailer: VM 6.75 under 21.1 (patch 10) "Capitol Reef" XEmacs Lucid X-face: "5Mnwy%?j>IIV\)A=):rjWL~NB2aH[}Yq8Z=u~vJ`"(,&SiLvbbz2W`;h9L,Yg`+vb1>RG% *h+%X^n0EZd>TM8_IB;a8F?(Fb"lw'IgCoyM.[Lg#r\ Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Stanley Hopcroft types: > Dear Ladies and Gentlemen, > . lpr ceased to work from the command line of an xterm (trace shows the > lpr client sending the P command but sending no data; interestingly > enough Netscape could still use lpr to print HTML with ghostcript). It > may well have also failed from the console. This one is *very* strange. My guess would be something in the environment was fried, but since it's fixed, there's not much you can do about it. > . reboot (several) shows normal device probes, file system checks but > then panics with > > init signal 6 exit 0 > panic: I'm not goiing anywhere without my init > rebooting in 10 seconds .... > > After booting the system with the live File System CD (thank you > CDROM.COM, thank you), and mounting the root file system of the defunct > 4.1-R system, I had a look at init. > > It was there and it looked fine. What do you mean by "looked fine"? Did you do a cmp on it and an offline copy that should have been ok? Did you save the init that wasn't working (always a good idea to do that kind of thing)? > . was there any relationship between the problem with lpr and the boot > failure (perhaps I ought to drag out the McKusick et al) Unknown, possibly unknowable. > . why on earth would init call abort() (the origin of the SIGABRT) ? It was probably damaged in some way. > . if init became "damaged", how could that happen - there's no sign of > problems reported by fsck / fsck doesn't check *data*, it checks *structure*. If something scrawls all over data blocks on your disk, fsck won't notice. > . is anything else likely to be "damaged". Should I reinstall - this is > a desktop, so the cost is not great ? I wouldn't. Some failures you *never* find an explanation for. These are sometimes called "cosmic ray failures", the implication being that a stray cosmic ray zapped a bit at the wrong time. Not necessarily what really happened, but the best you can do. > Has anywone in the history of computing known of MS systems being > repaired ? Yes, if you are sufficiently conversant with MS software, you can repair them instead of fix them. The level of training required for that is *much* higher than for FreeBSD, though.