Return to BSD News archive
Newsgroups: comp.os.386bsd.questions Path: sserve!newshost.anu.edu.au!munnari.oz.au!constellation!convex!convex!cs.utexas.edu!uunet!pipex!uknet!bradford.ac.uk!t.d.g.sandford From: t.d.g.sandford@bradford.ac.uk (Thomas Sandford) Subject: Re: NetBSD-0.9 partitioning Message-ID: <1993Sep2.103334.2646@info.brad.ac.uk> X-Newsreader: TIN [version 1.1 PL9] References: <MYCROFT.93Sep2031533@trinity.gnu.ai.mit.edu> Date: Thu, 2 Sep 93 10:33:33 BST Lines: 41 Charles Hannum (mycroft@trinity.gnu.ai.mit.edu) wrote: : In article <746948012.8813.0@unix7.andrew.cmu.edu> Christopher Dalton : <cd27+@andrew.cmu.edu> writes: : [Never done the first three, so I can't really comment on them.] : 4. Why can't NetBSD use the same install procedure as 386BSD : (which didn't require a clue)? : Why can't NetBSD lose as badly as 386BSD, you mean? The 386BSD : installation procedure created two partitions on my machine--one 5MB : swap partition, and the rest in root. In theory, this is a lousy : setup. In practice, it's even worse, and given that different systems : have different loads, it makes more sense to let people configure the : system for the load they expect. The NetBSD scheme is great for the *nix expert who *knows* what partition sizes they should be using. The 386BSD scheme was much more friendly to the Newbie (until they actually needed some swap :-} ). I would like to see an install that did something like: Number of sectors for sd0a partition [20480]: . Number of sectors for sd0b partition (recommend 2* ram size) [32768]: . etc. The default values being of an appropriate size, and taking into account the disk geometry to get integral cylinder multiples. Why should I have to do all the maths when the computer is *much* better at sums than I am? It might be harder to write such an install, but it would be *much* easier to use - and the experts could still tweak the values if they wanted. -- Thomas Sandford | t.d.g.sandford@bradford.ac.uk