[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Minutes of Debian Installer IRC meeting of 20040717

Minutes of Debian Installer IRC Meeting of 2004.07.17.2000

Dates and times are UTC-referenced.

Venue: #debian-boot at irc.debian.org

Meeting started at 20:00.

There were 107 IRC nicknames present at the start time.

The following people provided input to the meeting:
(a few names are missing as I cannot retrieve them currently)

 anibal		Anibal Monsalve Salazar
 anton		Anton Zinoviev
 AstinusUK	Alex Howells
 bdale		Bdale Garbee
 Bill_Jennings	Bill Jennings
 fs		Frederik Schueler
 GyrosGeier	Simon Richter
 huhlig		Hans W. Uhlig
 joeyh		Joey Hess
 joshk		Joshua Kwan
 kmuto		Kenshi Muto
 kyllikki	Vincent
 lieb		Jim Lieb
 luther		Sven Luther
 maswan		Mattias Wadenstein
 Md		Marco d'Itri
 mellum		Falk Hueffner
 Mithrandir	Tollef Fog Heen
 Q_		Kurt Roeckx
 seppy		Dennis Stampfer
 SimonRichter	Simon Richter
 SvenLuther	Sven Luther
 vorlon		Steve Langasek
 waldi		Bastian Blank


 aj		Anthony Towns
 bubulle	Christian Perrier
 calc		Chris Cheney
 cjwatson	Colin Watson
 dannf		Dan Frazier
 doogie		Adam Heath
 dsilvers	Daniel Silverstone
 gaudenz	Gaudezn Steiling
 fjp		Frans Pop
 jbailey	Jeff Bailey
 mckinstry	Alastair McKinstry
 pere		Petter Reinholdtsel
 tbm		Martin Michlmayr
 ths		Thiemo Seufer


 anibal		Anibal Monsalve Salazar

Discussion topics at:


Meeting log:
(please refer to the log to check who said what)


Next Release Date

joeyh suggested the end of the current month as next release date.
There was no objection.

Everybody would like to support 2.6 kernels. However there are some
issues preventing that:

* 2.4 is the only thing viable on sparc32 
* US and Amiga seem to have issues with 2.6 as well
* ia64 boot with 2.6 not ready yet 
* alpha won't be on 2.6 until until there's some sign of the PCI fixes
  making it into the kernel-images

String Freeze

A string freeze on general principle, so that people don't go changing
strings *after* everybody thinks they've got it all translated. There
are many of translated but un-uploaded strings.

To release by the end of the month, joeyh suggested a string freeze
starting in the next few days. There were no objections.

seppy will send the announcement to the translators.

Architecture Status


vorlon said that alpha is in pretty solid shape; the two big
outstanding issues for them are that automatic partitioning doesn't
yet account for the space aboot needs, and that 2.4 kernel images in
the archive are still in need of a major PCI fix. He estimates that 
it affects close to 1/4 of the alpha hardware. 2.6 hasn't been tried


Q_ said that amd64 currently doesn't have a netinst image. He has been
trying to get it working without much success. joeyh suggested Q_ to
talk to manty.

Q_ also said that they've also had some problems with changes in sid
base. It seems to work for some people while others say it failed
without much detail about why it failed. Some of those failures are
explained by sid base breakage because we only have sid.


kyllikki confirmed that there are no changes since last week when it
was reported in good shape.


jbailey has a hppa, and was hoping to test.

bdale and AstinusUK offered to test hppa.


joeyh reported that it seems to be in good shape. He also noted that
the lowmem patch has bugs, and doesn't know if zboob will get it fixed
in time.


lieb reported that they would like to get ia64 boot + 2.6 done before
final release. He also said that ia64 efi partman patches were submitted
this week. He's also re-writing the ia64 portions of the install manual.


joeyh has got confirmation that the genext2fs bug is still there. Also,
Florian Lohoff posted to -boot and reported the ext filesystem error
problem on boot.

Simon Richter will look after m68k/Amiga a bit.


joeyh said that someone reported on the list that the lowmem problem
with mips seems to not be a problem now.


SvenLuther said that Powerpc is in pretty good shape. The real problem
is that we really want 2.6 to be the default kernel on it, and this is
right now problematic.

He also said that he would make the 2.6 kernel the default, and have 2.4
as fallback (linux24). joeyh noted that at a minimum, he'll need to
offer both, joeyh can't see dropping 2.4 with 2.6 as of yet unproven.

SimonRichter APUS status: 2.4 doesn't boot, 2.6 doesn't build. However,
both seem trivial. His plan is to get both issues fixed by tuesday.


joeyh reported waldi and athornton are working on it. It mostly works
first stage, still need ssh for second stage.

waldi has done some work on ssh and it may be working in two weeks.


joshk said the best news is that SILO has purportedly been fixed.
Mail will be sent out to everyone who reported it failed.

The the smp kernel problem joshk mentioned last time has been worked
around. fs will test 2.6 on ultrasparc 1 smp.

d-i automation

The dumb way is to create a fictitious debconf priority above
critical, find a way to retrieve a debconf database and then raise
priority to fictitious one. The elegance of the priority system
(which, in the case of automation, could be considered a bug) is
that if there's an error you'll return to interactive mode.

Why not just use a custom backend? Say, http backend, all answers are
already seen, so they'll use the default answer. Change deconf.conf to
use the http backend, basically.


At present - many easy to fix bugs.  The real impression will come when
they are fixed. The coming week I will have enough time to work but
mostly offline. I will try not do do string changes after tuesday.

Meeting closed at 21:36.

Anibal Monsalve Salazar
 .''`.  Debian GNU/Linux      | Building 28C
: :' :  Free Operating System | Monash University VIC 3800, Australia
`. `'   http://debian.org/    | http://www-personal.monash.edu/~anibal/
  `-                          |

Attachment: signature.asc
Description: Digital signature

Reply to: