From owner-freebsd-questions@FreeBSD.ORG Fri Feb 17 23:23:23 2012 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 A406B106566B for ; Fri, 17 Feb 2012 23:23:23 +0000 (UTC) (envelope-from freebsd@edvax.de) Received: from mx02.qsc.de (mx02.qsc.de [213.148.130.14]) by mx1.freebsd.org (Postfix) with ESMTP id 51B328FC17 for ; Fri, 17 Feb 2012 23:23:23 +0000 (UTC) Received: from r56.edvax.de (port-92-195-127-244.dynamic.qsc.de [92.195.127.244]) by mx02.qsc.de (Postfix) with ESMTP id 396201DEE7; Sat, 18 Feb 2012 00:23:22 +0100 (CET) Received: from r56.edvax.de (localhost [127.0.0.1]) by r56.edvax.de (8.14.5/8.14.5) with SMTP id q1HNNLlR044994; Sat, 18 Feb 2012 00:23:21 +0100 (CET) (envelope-from freebsd@edvax.de) Date: Sat, 18 Feb 2012 00:23:21 +0100 From: Polytropon To: Devin Teske Message-Id: <20120218002321.55c738ad.freebsd@edvax.de> In-Reply-To: <021101ccedc9$89445cf0$9bcd16d0$@fisglobal.com> References: <4F3ECF23.5000706@fisglobal.com> <20120217234623.cf7e169c.freebsd@edvax.de> <20120217225329.GB30014@gizmo.acns.msu.edu> <021101ccedc9$89445cf0$9bcd16d0$@fisglobal.com> Organization: EDVAX X-Mailer: Sylpheed 3.1.1 (GTK+ 2.24.5; i386-portbld-freebsd8.2) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: freebsd-questions@freebsd.org Subject: Re: One or Four? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: Polytropon List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Feb 2012 23:23:23 -0000 On Fri, 17 Feb 2012 15:11:52 -0800, Devin Teske wrote: > > > > -----Original Message----- > > From: owner-freebsd-questions@freebsd.org [mailto:owner-freebsd- > > questions@freebsd.org] On Behalf Of Jerry McAllister > > Sent: Friday, February 17, 2012 2:53 PM > > To: Polytropon > > Cc: david.robison@fisglobal.com; freebsd-questions@freebsd.org > > Subject: Re: One or Four? > > > > > > > > > Let the majority decide which layout is preferred for the default. > > > > > > Why not add a selection to the installer, something like > > > this: > > > > > > Partition scheme > > > ---------------- > > > > > > [ ] all in one + swap > > > Create one partition containing all subtrees > > > plus one swap partition. > > > > > > [ ] separate partitioning + swap > > > Create /, /var, /tmp and /usr (including home) > > > partitions plus one swap partition. > > > > > > [ ] user-defined > > > Make your own partitioning selection manually. > > > > > > Of course, the default SIZES for second choice should be > > > reasonable. > > > > > > > Yes. Yes. This is the way to go. Let me change the caption of the second choice to this: [ ] traditionally separated partitioning + swap Because it's the installer's tradition to put /home into /usr. > I'd agree, but I'd like to envision a modular approach where multiple schemes > can be maintained. > > E.g. a menu containing... > > "Scheme 1: / + swap + /tmp" > "Scheme 2: / + swap + /tmp + /var" > "Scheme 3: / + swap + /tmp + /var + /usr" > "Scheme 4: / + swap + /tmp + /var + /usr + /home" I'm missing scheme 5 with /opt. :-) According to combinatoric possibilities, / + swap + /tmp + /usr is also missing. It would be no good idea (in my opinion) to present the user a list of _all_ possible combinations just in case he would like to have one of them. My idea to use three options ("minimal", "traditional", user-defined) would be fully sufficient, as all those who have no idea what they do would use the first choice, those who intendedly want the traditional approach would use the second choice, and all those not wanting one of those would be clever enough to deal with manually defining their own scheme. > I'm actually thinking that "not having a separate /tmp" is: > > a. A security issue > > /tmp is by-default out-of-the-box world-writable (perms 1777). Making this > world-writable bucket part of "/" seems silly both for Desktops and Servers > alike. Fully agree. I pointed out why this can be dangerous. Having /tmp in memory is good (and secure!) if it's possible (note: enough RAM needed), but not an option on systems low on RAM. This kind of possible fine tuning partition-wise (soft updates, journaling, quota, dump, ro, noexec, noatime etc.) doesn't typically take place on average desktops, but there may be cases where you need to do that. > b. A nuisance > > As "Da Rock" points out, ... recovering your system from a > file-system-full-event when using "single-/" is just as difficult regardless of > Desktop versus Server. Having "/tmp" alleviates the difficulty. I don't think the separation "desktop vs. server" serves well here. It's not about what kind of machine (or form factor) is used, but the actual _employment_ of the machine, the intended way of using it is. Note that there are also mixed forms, e. g. a home desktop that provides some server functionalities. That's why I think making a selection for partitioning schemes should take SCHEMES into mind, not "server or desktop". > c. A performance issue > > I'm surprised nobody has pointed out the physical performance limitations of > rotating disks with respect to physical location of partitions on the spindle. > Granted, seek times are light years beyond what they used to be, but allocating > smaller "swap" and "tmp" partitions close to the center of the spindle is a > performance-enhancing setup just as much as it is for protecting against > file-system-full problems (security events included). As I said, sometimes you need to squeeze every bit of performance out of a machine. Fiddling with the location of certain functional pieces of the OS _on the disk_ can be a big help here. > I'd argue that there should never be a single-"/" unless you are prepared to > deal with a truly 100%-full filesystem problem (especially considering that > Desktop users whom select the default-everything are often not skilled enough to > deal with that situation). If someone truly wants a single "/" and nothing else, > there's manual partitioning (which should prove pretty easy in the event that > you're only creating one partition and nothing else). Yes, that's also possible, but I think having it as a option to be checked is what especially novice users would want. They would select the first (default) choice anyway without reading, so it might be a chance to learn. :-) -- Polytropon Magdeburg, Germany Happy FreeBSD user since 4.0 Andra moi ennepe, Mousa, ...