From owner-freebsd-questions Fri Sep 12 01:51:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id BAA18152 for questions-outgoing; Fri, 12 Sep 1997 01:51:10 -0700 (PDT) Received: from gate.mgt.msk.ru (mgtrep.24h.dialup.ru [194.87.18.139]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id BAA18121 for ; Fri, 12 Sep 1997 01:50:51 -0700 (PDT) Received: from asteroid.mgt.msk.ru (asteroid.mgt.msk.ru [192.168.133.145]) by gate.mgt.msk.ru (8.8.6/8.8.6) with ESMTP id MAA02313 for ; Fri, 12 Sep 1997 12:49:44 +0400 (MSD) Received: from asteroid.mgt.msk.ru (localhost.mgt.msk.ru [127.0.0.1]) by asteroid.mgt.msk.ru (8.8.6/8.8.6) with ESMTP id MAA00907 for ; Fri, 12 Sep 1997 12:50:04 +0400 (MSD) Message-Id: <199709120850.MAA00907@asteroid.mgt.msk.ru> To: "'freebsd-questions@freebsd.org'" Reply-To: tarkhil@mgt.msk.ru Subject: Re: Clustering/fail-over capability? In-reply-to: Your message of "Wed, 10 Sep 1997 19:55:32 PDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 12 Sep 1997 12:50:04 +0400 From: "Alexander B. Povolotsky" Sender: owner-freebsd-questions@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > Is there anything built-in to FreeBSD similar to allow us to set up an > > identical secondary server and when the primary server stops > > communicating, have it take over (with the failed servers IP addresses) > > and try and shutdown the primary server? > system to do this nicely and cleanly. However, doing automatic takeover > isn't done nicely and cleanly. You have to define what is meant by `stops > communicating', and how to remotely shut down a machine without creating a > gigantic security hole in the process. (If you can do this securely, the > UPS people want to talk to you.) We're thinking on the same thing here, and we're going to monitor needed services regularry, and on shut of them try to shutdown the whole server, than tell UPS to turn it off, than... I think we should just change config files and kill -HUP some daemons, redo ifconfigs and start needed services... my co-worker says we need to change fstab and boot with complete mirror. We're not going to do 100% mirroring, making mirror once a day instead. Since one stupid but expencive program didn't find a license and destroyed 20 Gb of data at once, we're kinda concerned on these things. Alex.