From owner-freebsd-chat Wed Aug 6 17:23:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA00287 for chat-outgoing; Wed, 6 Aug 1997 17:23:48 -0700 (PDT) Received: from nico.telstra.net (nico.telstra.net [139.130.204.16]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id RAA00253; Wed, 6 Aug 1997 17:23:39 -0700 (PDT) Received: from freebie.lemis.com (gregl1.lnk.telstra.net [139.130.136.133]) by nico.telstra.net (8.6.10/8.6.10) with ESMTP id KAA09276; Thu, 7 Aug 1997 10:24:02 +1000 From: Greg Lehey Received: (grog@localhost) by freebie.lemis.com (8.8.7/8.6.12) id JAA01467; Thu, 7 Aug 1997 09:53:00 +0930 (CST) Message-Id: <199708070023.JAA01467@freebie.lemis.com> Subject: Re: Hot Swappable Kernels In-Reply-To: <199708070004.TAA00279@argus.tfs.net> from Jim Bryant at "Aug 6, 97 07:04:16 pm" To: jbryant@tfs.net Date: Thu, 7 Aug 1997 09:53:00 +0930 (CST) Cc: hackers@freebsd.org (FreeBSD Hackers), chat@freebsd.org (FreeBSD Chat) Reply-To: chat@freebsd.org (FreeBSD Chat) Organisation: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8250 Fax: +61-8-8388-8250 Mobile: +61-41-739-7062 WWW-Home-Page: http://www.lemis.com/~grog X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-chat@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Jim Bryant writes: > In reply: >> Jamil J. Weatherbee writes: >>> >>> I know this may sound kind of lame: >>> >>> I was thinking last night of what would be required to have a hot >>> swappable kernel.. i.e. being able to compile the kernel binary (probably) >>> modules and then insert it into a running system while maintaining its >>> running status --- to my knowledge kernel recompiles are the only reason a >>> perfectly rebooting system needs to come down every once in a while. >> >> I suspect people are going to shoot you down in flames, and they're >> probably justified. But I suppose you'd like to know that I've done >> just that in the past, at Tandem. The operating system is a loosely >> coupled network, so we were able to boot one machine at a time. >> Despite the obvious interest of such a scheme for Tandem, and despite >> my extensive lobbying, it never came to anything. >> >> I can't imagine how you would start to do such a thing with UNIX. The >> closest you could come to it would be to split most of the kernel into >> LKMs, and change them. But there's a basic conflict of concept >> between keeping a kernel running (even if it's no longer the same >> kernel) and booting a kernel. > > greg, > > Keep in mind that even for Guardian-90, you still have to boot... Sure, but with my method, you booted one CPU at a time. So the system really didn't go down. > Let's see, what changes if i simply change MAXUSERS... > > This is one of the simpler scenarios... Changing other things can be > really hairy... Sure, I'm not arguing that it should be possible to change a number of parameters without rebooting. On the contrary, it's a good idea. But it's a long way from there to swapping kernels on the fly. > Unix on a PC is not Guardian-90 on a Non-Stop. An occassional reboot > is not a bad thing. An occasionaly reboot is a necessary evil. That's just another way of saying that it's a Bad Thing. > Prediction: Three years, tops. Compaq will run Tandem into the > ground. A shame too... Hmm. I've spoken to people at Tandem, and the opinion is mixed. Remember Tandem's been running itself into the ground for a while. BTW (and thus the follow-up to -chat?): what are they going to call the new company? Tampaq or Condem have been suggested. Greg