Re: The size of debian packages
Andrew Lyon <[EMAIL PROTECTED]> writes:
> Since installing Debian in May the amount of disk space required for
> my OS has risen from 1.5GB to 3GB and has reached the limit of the
> partition. I don't really want to allocate any more space to the OS
> as I'm sure there must be stuff there that I do not need or use (I
> admit to running dselect a little to liberally in the past!). How
> can I find out the sizes of the packages and try to establish what I
> can remove without disaster. I tried using deborphan to do this but
> it didn't even put a dent in my 100% full volume.
You might try debfoster. synaptic makes it pretty easy to see your
installed packages, sorted in order of size.
This script shows installed packages together with their installed
size:
#!/bin/sh
gawk 'BEGIN {RS=""; FS="\n"}
$2~/ installed/ {
split($1, pkg, " ");
split($5, size, " ");
print pkg[2], size[2];}' /var/lib/dpkg/status
I call it "installed". Then things like "installed | sort -n -k 2 |
less" can be useful.
[...]
Re: Bug #213524
Jochen Friedrich <[EMAIL PROTECTED]> writes: > bug #213524 on automake has the potential to hit a lot of packages > in a way that they become uninstallable. So far, diff, indent and > nano have been affected, but there might be more to come. Ah! I've recently noticed that a recent install didn't seem to have odd info files missing from /usr/share/info/dir (in some cases even though I'd installed the package that day). This bug sounds like it's the one responsible. (Well, the change in behaviour of Debian's install-info, anyway.) > What could be done to limit the damage? No idea. It's annoying, however. Much more serious if packages are actually uninstallable, of course, although I haven't noticed that yet. Is there some way to rebuild /usr/share/info/dir?
Re: Are we losing users to Gentoo?
Jon Kent <[EMAIL PROTECTED]> writes: [...] > What we need to accept is there is a (percieved??) > problem, or problems, with Debian as it stands today, > these being (mainly) > > Hard to install (rubbish obviously) > Out of date (this _is_ true) > Slow to update (this _is_ true) > Hard to configure (depends upon your view-point) Releases tend to be out of date. But that's a feature: releases need to be composed of well tested stable packages. testing and unstable have pretty up to date packages. So Debian is as up to date as you want; the caveat being that for newer software, you'll need to put up with some instability. > The reasons I see people switch to Gentoo are : > > Its more fun > Alot more up to date > Easier to customise, down to which libraries you want > to support Presumably its up to dateness comes at the cost of less stability? So probably people should compare Gentoo not with Debian releases (stable), but with testing (or perhaps even unstable)? How do they compare then? > Gentoo is still hard to configure if you are only used to Red Hat or > Mandrake, easy if you used to Debian, Slackware etc. > > IMHO Debian is too slow to put out new releases. I > run testing to ages with no problems, ever. Sure on > my unstable box things went south at times but I > expect that and can fix it, but testing is very solid, > as solid as, say, Red Hat. Yes, possibly. Quite a bit of the problem seems to come with preparing boot floppies, of all things. Maybe there's some case for making a regular (once every couple of months or so) "State of Testing" announcement, describing the major features of testing, together with how to install it (either "install stable release, then change /etc/apt/sources.list thusly, then do "apt-get update; apt-get dist-upgrade", or perhaps actually preparing a Knoppix ISO containing testing). On the other hand, maybe this wouldn't be much use to anyone. [...]
Re: plagiarism of reiserfs by Debian
Andreas Dilger <[EMAIL PROTECTED]> writes: [...] > I'm sure all the FSF/Debian folks would be thrilled if someone > changed the code in [x]emacs to not output anything about the GPL at > startup, or if vim didn't include any info about helping Ugandan > orphans. XEmacs and Emacs follow sensible guidelines: if you start them without any arguments, then they display useful information, including information about the GNU GPL. If you start them such that they can usefully display something else (like a filename that you want to edit), then they don't display the other information, and you don't get reminded about the GNU GPL. So the information isn't thrust in anybody's face: if you don't give it any arguments, Emacs doesn't have anything else it might display, so it may as well display information about itself (how to get help, conditions of copying, etc.). To remove this code would be a bad technical decision---there's no reason to. I presume (if any code has been changed) that some of the reiserfs code is doing something that's less technically justifiable. (I don't know about vim.) [...]
Re: sid: libc6-2.2.5-4 kills vmware workstation 3.0
"Bao C. Ha" <[EMAIL PROTECTED]> writes: > On Mon, Apr 08, 2002 at 11:03:11AM -0500, Donald J Bindner wrote: > > Hi Donald, > >> >> Let me see if I understand this. I am running VMWare 2.0.4 and >> this morning I discovered that it dies with: >> >> VMware Workstation PANIC: >> AIO: NOT_IMPLEMENTED F(566):1081 >> >> This is on a relatively current Woody system, and VMWare was >> running fine last week. Is this the same issue, and does that >> leave me in the "sorry" category? > > It is the same issue. > > You will need to get the latest patch from Petr, > > ftp://platan.vc.cvut.cz/pub/vmware/vmware-ws-any-update14.tar.gz > > It fixes the problem. I'd guess an alternative would be a simple LD_PRELOAD trick to override getpriority/setpriority, with the caveat that such tricks presumably don't work on setuid programs, so you'd have to be running as root? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
