From owner-freebsd-questions@FreeBSD.ORG Sun Nov 4 07:56:18 2007 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id EAEFC16A469 for ; Sun, 4 Nov 2007 07:56:18 +0000 (UTC) (envelope-from chris@sigd.net) Received: from ms05.mailstreet2003.net (ms05.mailstreet2003.net [69.25.50.235]) by mx1.freebsd.org (Postfix) with ESMTP id B2EFB13C4A8 for ; Sun, 4 Nov 2007 07:56:18 +0000 (UTC) (envelope-from chris@sigd.net) X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Date: Sun, 4 Nov 2007 02:56:04 -0500 Message-ID: <6FC9F9894A9F8C49A722CF9F2132FC220F34B36D@ms05.mailstreet2003.net> In-Reply-To: <44sl3m7ogj.fsf@Lowell-Desk.lan> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: 5.2.1 to 6.2 Migration. Thread-Index: AcgemM0oV8pxJWLLTLmfpJ+IpeXB1AAJ3nTw References: <011e01c81e3a$82335dc0$6501a8c0@GRANT><6FC9F9894A9F8C49A722CF9F2132FC220F34B2E6@ms05.mailstreet2003.net><44640jrpsp.fsf@Lowell-Desk.lan> <472D002F.10603@daleco.biz> <44sl3m7ogj.fsf@Lowell-Desk.lan> From: "Chris Haulmark" To: Subject: RE: 5.2.1 to 6.2 Migration. X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 04 Nov 2007 07:56:19 -0000 >=20 > Kevin Kinsey writes: >=20 > > Lowell Gilbert wrote: > >> "Chris Haulmark" writes: > >> > >>> Grant Peel wrote: > > > >>>> I thought I would ask the question before I do it the hard way > .... > >>>> > >>>> 1. Can FreeBSD be upgraded from 5.2.1 to 6.2 ? > > > >>> Yes. > > > >>>> 2. Can it be done through an ssh connection, or MUST I make the > trip > >>>> to the farm and do it from the console? > > > >>> I've done 5.x to 6.x upgrades via ssh. It is possible. > >>> > >>> In the handbook, you will see mentions of booting into single user > mode > >>> and I can tell you that it is not required. > >> > >> It's a good safety precaution; if your updated kernel won't boot, > you > >> will need to reinstall most of the system. That is over the board. =20 Only times that I have made the mistakes in the past are: 1. Misconfiguring the kernel options such as disabling the meeded network driver built in the kernel. 2. Anything related to having kernel panics to occur. 3. Enabling firewall and getting locked out via network. > > > > That sounds a tad alarmist; if the new kernel won't boot, you'll > > have to be at (or have someone at) the console who can boot > > "kernel.old" (I stand open for correction, but last time I did > > it, 'twas that way). And, possibly, that person (you?) will > > also have to be able to do some other magic. Magic such as having other remote possibilities. DRAC access for example. > > > > But the phrase "reinstall most of the system" doesn't, at > > the very least, *sound* like the BSD Way(tm). Granted, > > sometimes it's quicker --- I know that's why it's used so > > often on that "Other System" .... ;-) >=20 > If you have reinstalled a userland that depends on a kernel that > doesn't boot, you are quite likely to be in trouble. I always do buildworld/installworld as part of my kernel build/installs. That is to ensure staying in sync. I reboot after the installworld then again after the installkernel. >=20 > The "BSD way" does not necessarily involve easy recovery from making > up procedures that haven't been worked out or tested by the release > engineers. In fact, I don't think any operating system guarantees > that you will have an easy time after making up your own upgrade > procedures.