Date: Sun, 1 Feb 1998 12:54:14 +1030 From: Greg Lehey <grog@lemis.com> To: Malte Lance <malte@webmore.com> Cc: questions@FreeBSD.ORG Subject: Re: system unstable, how to make it stable again ? Message-ID: <19980201125414.10342@lemis.com> In-Reply-To: <3.0.32.19980201025749.02f83300@cyclone.degnet.baynet.de>; from Malte Lance on Sun, Feb 01, 1998 at 02:57:49AM -0100 References: <3.0.32.19980201025749.02f83300@cyclone.degnet.baynet.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Feb 01, 1998 at 02:57:49AM -0100, Malte Lance wrote: > At 09:41 01.02.98 +1030, you wrote: >> On Sat, Jan 31, 1998 at 10:56:22AM -0100, Malte Lance wrote: >> >>> I have disklabeled / and /usr with disktab-entries that are >>> parially wrong. That never caused any problems so i was always >>> too lazy to correct it. And now i hesitate to update a broken >>> system. I am considering a fresh install to be the better way to >>> go. On the other hand i have installed many ports and configured >>> the machine to be mail-hub, NS, www, caching-proxy and >>> firewall. It would take many days of configuration-work to get a >>> fresh install to the same point. >> >> Not really. Just read in the backup. > > Right. > BTW ... does dump restore file-pipes and devices correct ? I don't know. I don't use dump. tar restores files with small minor numbers. I don't think you need to restore named pipes: the ones the system uses are created at boot time. >>>> You've got a lot of peripherals on this machine. Can you see any >>>> connection between peripheral activity and the hangups? >>> >>> Nope. The funny thing is, when the machine freezes, the harddrives, >>> when accessed at this moment, dont even have time to switch the lights off. >> >> Ah! There's your answer! It's a SCSI bus problem. I thought it >> might be. Are you sure you didn't add any peripherals recently? > > Yes ... besides the DAT-Streamer. That's enough. Presumably it's DDS, BTW. >>> No errors from the SCSI-system. >>> A few days ago i replugged the Sony-DAT but did not use it so far, just one >>> tar-backup on /usr >> >> OK. There's your problem (I hadn't read this far when I made my >> previous comment). I'd suspect termination. > > No, it's in the chain and termination is disabled. OK. To quote: SCSI is *NOT* magic. There are *fundamental technical reasons* why it is necessary to sacrifice a young goat to your SCSI chain now and then. --jfw@proteon.com (John Woods) >> Ah. Don't expect a 'make world' to cure problems. I think Doug White >> has some documentation on updating, but I don't know where. We > > Doug, are you reading this thread ? > Would you please give me the link Greg is talking about. > TIA > > I give it two more crashes ... then i'll update :) Why not disconnect the DDS drive after the next hang? Greg
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19980201125414.10342>