Archive consistency stuff

[Anthony Towns <aj@azure.humbug.org.au>]

Hello world,

Some more archive consistency stuff.

First, the number of out of date packages in potato. (These are binaries
that seem to be based on a different source version to the one that's
actually in potato, not counting -x.y.[0-9]* binary only arch specific
reuploads)

      42    1 i386
     101    9 alpha
     170   31 sparc
     203   13 arm
     238    9 m68k
     333   10 powerpc

The first column is the number of binary packages, the second is the
number of binary packages whose binaries are newer than the source. Note
that some of these are binary only NMUs (ie -x.[0-9]*) which are probably
just mis-versioned binary only recompiles, and others are binary only
recompiles of Debian-native packages, which aren't easily detectable.

For woody:

      79    3 i386
     124    0 hurd-i386
     428    5 alpha
     463    9 m68k
     540   17 arm
     622   29 sparc
     671   10 powerpc

For slink:

      62    6 i386
      95   18 m68k
     133   73 alpha
     264  224 sparc

For testing:

      61    3 i386
     113    6 alpha
     123   34 sparc
     160   12 arm
     186    9 m68k
     339   19 powerpc

The full reports should be available, updated daily, at:
	http://auric.debian.org/~ajt/slink_outdate.txt
	http://auric.debian.org/~ajt/potato_outdate.txt
	http://auric.debian.org/~ajt/testing_outdate.txt
	http://auric.debian.org/~ajt/unstable_outdate.txt

Packages marked with *** are the ones that seem to be newer than their
source. The summaries are right at the bottom. Note that this is probably
timed badly, in that it probably doesn't count anything that's been
uploaded for the last couple of days.


Okay, next interesting thing is how woody's been messed up after the
freeze. Things like gcc being removed, and libc6's disappearance from
alpha and arm and such. These are, apparently, due to people uploading
binaries to frozen (only) and thus making the symlink from unstable
disappear (where the frozen and unstable versions haven't yet diverged).

The following sources have apparently suffered this fate:

source:
    ace 5.0.7-2
    blacs-mpi 1.1-9
    blacs-pvm 1.1-7
    bridgex 0.30
    gcc 2.95.2-10
    gdb-arm 200000308-1
    gnap 0.1.5-3
    ivtools 0.7.9-6
    kernel-source-2.0.38 2.0.38-3
    libgc4 1:4.14-7
    libpng0 0.96-5
    nfs-server 2.2beta47-4potato.2
    nsmon 2.3e-3
    pdl 2.005-2
    prc-tools 0.5.0r-3.1
    sam 4.3-9
    task-chinese 0.6
    wmf 1.0.5-3
    workman 1.3.4-3
    ytree 1.65-4

And the following (additional) i386 binaries have also:

i386:
    alsa-base 0.4.1i-4
    alsa-headers 0.4.1i-4
    alsa-source 0.4.1i-4
    alsalib0.3.0 0.4.1e-2
    alsautils 0.4.1-5
    dict-jargon 1.5-2
    icecast-client 1.0.0-1
    icecast-server 1.0.0-1
    iczech 19990918-1
    libasound0.4 0.4.1e-2
    libasound0.4-dev 0.4.1e-2
    libg++2.8 2.90.29-2
    libg++2.8.2-glibc2.1 2.91.66-4
    libreadlineg2 2.1-19
    python-netcdf 1.03-3

All these can nominally be fixed just be adding symlinks from
woody. Theoretically, anyway. Some of them may be deliberate, of course.
This may not be all the packages that have suffered a similar fate either,
they're just most of the ones that seem to be easlily fixable by adding
a symlink.

Full stats, again updated daily, are at:
	http://master.debian.org/~ajt/potato-not-woody.txt

Cheers,
aj

-- 
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG encrypted mail preferred.

  ``We reject: kings, presidents, and voting.
                 We believe in: rough consensus and working code.''
                                      -- Dave Clark

Attachment: pgpn8mU0q5VZx.pgp
Description: PGP signature