Re: major linux problems summary 2012

[Tollef Fog Heen]

]] "Andrej N. Gritsenko" 

> That is no case for open software as you still have no money from it,
> whether would you fix the bugs or not, therefore will someone use it
> beside you or not, it still will be for your own fun.

You seem to be confusing «open software» (sic) with non-commercial
software.  They are not at all the same.

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87a9uxwxf2@xoog.err.no

Re: IPv6, tentative addresses, bind(), wheezy

[Marc Haber]

On Sat, 3 Nov 2012 00:38:32 +0100, m...@linux.it (Marco d'Itri) wrote:
>On Nov 02, Don Armstrong  wrote:
>> Seems like the right solution is to wait for the network device to be
>> fully up in the init script with a sleep loop... though that's
>> certainly not optimal.
>Maybe for the time being it will be easier and safer to have ifupdown 
>wait until DAD is finished?

How does one find out about that?

Greetings
Marc
-- 
-- !! No courtesy copies, please !! -
Marc Haber |   " Questions are the | Mailadresse im Header
Mannheim, Germany  | Beginning of Wisdom " | http://www.zugschlus.de/
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 621 72739834


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/e1tuwgl-0004zl...@swivel.zugschlus.de

Re: IPv6, tentative addresses, bind(), wheezy

[Marc Haber]

On Fri, 2 Nov 2012 23:24:55 +0100, martin f krafft
 wrote:
>The reason is that by the time bind() is called, the IPv6 address
>(configured with /e/n/i inet6 static, which unbound should listen
>on) is not yet ready, but "tentative", so the bind() call fails.

Last time when I asked a question along those lines (with bind as the
application), the answer was the usual Debian-like "your application
is broken, have upstream convert the app to an event-based approach
and do not bother us".

This answer is partially right, applications need to be able to handle
dynamically changing IP addresses at run-time to properly support
IPv6. If they don't, one of the major advantages of IPv6 ("renumbering
is easy") is lost. This goes especially if the app's configuration
contains IP addresses.

The problem is, that the distributions, IMO, need to work around these
shortcomings until the applications are eventually fixed (which might
never happen). The socket API also has several shortcomings which
makes this approach harder to do. For examle, there is no semantic for
operations like "listen on this host address in all prefixes that the
host system is aware of", probably written along the line of
::224:d7ff:fed0:5adc.

Greetings
Marc
-- 
-- !! No courtesy copies, please !! -
Marc Haber |   " Questions are the | Mailadresse im Header
Mannheim, Germany  | Beginning of Wisdom " | http://www.zugschlus.de/
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 621 72739834


--
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/e1tuwlx-00050r...@swivel.zugschlus.de

Re: major linux problems summary 2012

[Andrey Rahmatullin]

On Sat, Nov 03, 2012 at 02:37:34PM -0700, Russ Allbery wrote:
> At least with open source software, a third path (besides working around
> the bug or switching to some other software) also exists: you can help.
> That's not always practical, but at least it's *possible*.
Not all problems are things that can be fixed by applying a text editor.
Some of them are social, some would require major overhauls of major
ecosystem components, probably in ways that will be rejected by current
developers.

-- 
WBR, wRAR


signature.asc
Description: Digital signature

Re: major linux problems summary 2012

[Andrey Rahmatullin]

On Sat, Nov 03, 2012 at 10:29:50PM +0100, Michael Banck wrote:
> If there are specific issues which cannot be expressed by bug reports
> (e.g. because they affect more than one package), by all means raise
> them here one by one.
"You can't publish a (dynamically linked) binary that can be used on most
user systems"
"You cannot make all your GUI software look the same"

-- 
WBR, wRAR


signature.asc
Description: Digital signature

Re: major linux problems summary 2012

[Karol Szkudlarek]

2012/11/3 Jonathan Nieder :
> Hi Karol,
>
> Karol Szkudlarek wrote:
>
>> I'm Debian/Ubuntu/Linux enthusiasts which loves using free operating
>> systems on my machines, but today in 2012 I'm struggling with many
>> problems with setup/tunning my Dell Latitude E6510 and Ubuntu and I
>> have some thoughts general thoughts about Linux lacking and problems.
>>
>> And in the internet I've found great summary of my thougths written by
>> Artem S. Tashkinov about Linux:
>>
>> http://linuxfonts.narod.ru/why.linux.is.not.ready.for.the.desktop.current.html
>
> I've found you to be helpful and constructive when dealing with kernel
> bugs, so I don't want you to be discouraged.
>
> A ping listing bugs that need more attention or a new strategy to solve
> them is not a bad thing.  However, the above list is not presented in a
> helpful way --- it would be more useful to list bug numbers and to be
> specific when possible about what direct action the reader can take to
> make things better.
>
> For example:
>
> "I have been struggling with many problems setting up and
>  tuning my Dell Latitude E6510 on Ubuntu that are also present
>  in Debian:
>
>  bug#12345 - ...
>  bug#12346 - ...
>
>  The process has been slow and discouraging.  Any tips for
>  making this go better?  What can I and my friends do to make
>  sure the next new machine is closer to working on Debian out
>  of the box without fuss?"
>
> By the way, the kernel team needs help --- many of the bugs at
> http://bugs.debian.org/src:linux-2.6 and http://bugs.debian.org/src:linux
> need the work listed at
> http://kernel-handbook.alioth.debian.org/ch-bugs.html#s9.1.4 done for
> them.
>
> Thanks and hope that helps,
> Jonathan

Hello Jonathan,

Thanks for explantation and of course I'm trying to fill bug reports
to make things better..
Currently my laptop works under Ubuntu and here are examples of BTS reports:

about BUG: scheduling while atomic:

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1068833

and about graphics freeze:

https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers/+bug/1033263

and about touchpad:

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/606238

and about cpu fan:

https://bugs.launchpad.net/ubuntu/+source/i8kutils/+bug/410596

https://bugs.launchpad.net/ubuntu/+source/acpi/+bug/632228

https://bugs.launchpad.net/ubuntu/+source/i8kutils/+bug/1060096

I don't know where are also Debian specific, because I'm using my
laptop to day by day
work with many stuff configured as I want so changing/replacing kernel
or a distro is a
process which I can't do frequently (probably first I will try a live
demo of wheezy when will
be released)

All I need is a stable set of tools and mobile laptop which I can
resume/suspend without
crashes and do my work. Secondly I'm only a user of OS not a developer. :-)

I'm monitoring crashes of my laptop which can't be solved during
Ctrl-Alt-Backspace or REISUB and
the stats from April look as follows:

crash 2012-04-29 11:27:26 -> up 2012-04-29 13:46:20 =02:18:54 (8334 s)
crash 2012-05-06 12:11:10 -> up 2012-05-06 12:28:24 =00:17:14 (1034 s)
crash 2012-05-07 16:31:23 -> up 2012-05-07 21:35:08 =05:03:45 (18225 s)
crash 2012-05-07 21:37:54 -> up 2012-05-07 21:41:04 =00:03:10 (190 s)
crash 2012-05-07 21:47:29 -> up 2012-05-07 21:48:00 =00:00:31 (31 s)
crash 2012-05-14 21:44:42 -> up 2012-05-14 23:11:07 =01:26:25 (5185 s)
crash 2012-06-07 22:42:24 -> up 2012-06-08 08:05:23 =09:22:59 (33779 s)
crash 2012-06-27 19:00:23 -> up 2012-06-27 19:25:18 =00:24:55 (1495 s)
crash 2012-06-27 20:48:10 -> up 2012-06-27 22:06:09 =01:17:59 (4679 s)
crash 2012-06-30 18:03:12 -> up 2012-06-30 18:04:06 =00:00:54 (54 s)
crash 2012-07-20 22:26:26 -> up 2012-07-20 22:29:49 =00:03:23 (203 s)
crash 2012-08-07 21:05:38 -> up 2012-08-07 21:19:29 =00:13:51 (831 s)
crash 2012-08-07 22:37:09 -> up 2012-08-07 22:40:40 =00:03:31 (211 s)
crash 2012-08-12 17:36:20 -> up 2012-08-12 18:20:58 =00:44:38 (2678 s)
crash 2012-08-16 10:27:22 -> up 2012-08-16 10:33:18 =00:05:56 (356 s)
crash 2012-08-21 19:37:52 -> up 2012-08-21 19:38:14 =00:00:22 (22 s)
crash 2012-09-01 23:14:06 -> up 2012-09-01 23:17:22 =00:03:16 (196 s)
crash 2012-09-19 10:53:17 -> up 2012-09-19 10:55:38 =00:02:21 (141 s)
crash 2012-09-22 17:53:34 -> up 2012-09-22 17:59:28 =00:05:54 (354 s)
crash 2012-09-23 17:47:52 -> up 2012-09-24 08:24:35 =14:36:43 (52603 s)
crash 2012-09-30 21:05:59 -> up 2012-09-30 21:23:52 =00:17:53 (1073 s)
crash 2012-10-01 20:39:22 -> up 2012-10-01 20:47:07 =00:07:45 (465 s)
crash 2012-10-05 23:52:09 -> up 2012-10-06 09:46:29 =09:54:20 (35660 s)
crash 2012-10-08 21:53:00 -> up 2012-10-08 21:54:57 =00:01:57 (117 s)
crash 2012-10-11 23:09:10 -> up 2012-10-12 09:38:53 =10:29:43 (37783 s)
crash 2012-10-13 06:33:47 -> up 2012-10-13 13:15:08 =06:41:21 (24081 s)
crash 2012-

Re: major linux problems summary 2012

[Michael Banck]

On Sun, Nov 04, 2012 at 04:01:03PM +0600, Andrey Rahmatullin wrote:
> "You can't publish a (dynamically linked) binary that can be used on most
> user systems"
> "You cannot make all your GUI software look the same"

The information density and/or applicability of those statements is too
low to usefully discuss them in the context of debian-devel, sorry.


Michael


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20121104114113.gb20...@nighthawk.chemicalconnection.dyndns.org

off topic (Re: major linux problems summary 2012)

[Bart Martens]

On Sun, Nov 04, 2012 at 12:03:37PM +0100, Karol Szkudlarek wrote:
> > Karol Szkudlarek wrote:
> >> today in 2012 I'm struggling with many
> >> problems with setup/tunning my Dell Latitude E6510 and Ubuntu
> Currently my laptop works under Ubuntu
> I don't know where are also Debian specific
> Secondly I'm only a user of OS not a developer. :-)

This mailing list is about "development of Debian", not about "using Ubuntu".
http://lists.debian.org/debian-devel/

In case you want to try using Debian, here's a mailing list for "help and
discussion among users of Debian":
http://lists.debian.org/debian-user/

Regards,

Bart Martens


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121104115056.gb6...@master.debian.org

Bug#692264: ITP: double-conversion -- routines to convert IEEE doubles to and from strings

[Sébastien Villemot]

Package: wnpp
Severity: wishlist
Owner: "Sébastien Villemot" 
User: debian-scie...@lists.debian.org
Usertags: field..mathematics
Control: block 691912 by -1

* Package name: double-conversion
  Version : 1.1.1-1
  Upstream Author : Florian Loitsch 
* URL : http://double-conversion.googlecode.com
* License : BSD
  Programming Lang: C++
  Description : routines to convert IEEE floats to and from strings

The library provides routines to convert IEEE single and double floats
to and from string representations. It offers at lot of flexibility with
respect to the conversion format: shortest, fixed, precision or
exponential representation; decimal, octal or hexadecimal basis; control
over number of digits, leading/trailing zeros and spaces.

The library consists of efficient conversion routines that have been
extracted from the V8 JavaScript engine. The code has been refactored
and improved so that it can be used more easily in other projects.

-- 
 .''`.Sébastien Villemot
: :' :Debian Developer
`. `' http://www.dynare.org/sebastien
  `-  GPG Key: 4096R/381A7594


pgpdkYEePxhEr.pgp
Description: PGP signature

Re: Bug#692264: ITP: double-conversion -- routines to convert IEEE doubles to and from strings

[Roger Leigh]

On Sun, Nov 04, 2012 at 12:55:09PM +0100, Sébastien Villemot wrote:
> 
> * Package name: double-conversion

libjs-double-conversion would be more appropriate, since
this is a javascript library.  double-conversion is too
generic.


Regards,
Roger

-- 
  .''`.  Roger Leigh
 : :' :  Debian GNU/Linuxhttp://people.debian.org/~rleigh/
 `. `'   schroot and sbuild  http://alioth.debian.org/projects/buildd-tools
   `-GPG Public Key  F33D 281D 470A B443 6756 147C 07B3 C8BC 4083 E800


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121104120327.gg14...@codelibre.net

Re: Bug#692264: ITP: double-conversion -- routines to convert IEEE doubles to and from strings

[Sébastien Villemot]

Roger Leigh  writes:

> On Sun, Nov 04, 2012 at 12:55:09PM +0100, Sébastien Villemot wrote:
>> 
>> * Package name: double-conversion
>
> libjs-double-conversion would be more appropriate, since
> this is a javascript library.  double-conversion is too
> generic.

It's actually written in C++, not JS, so I don't think the libjs prefix
is appropriate.

The name "double-conversion" is taken from upstream, but I agree that it
is quite generic. I don't have any better idea, but suggestions are
welcome.

-- 
 .''`.Sébastien Villemot
: :' :Debian Developer
`. `' http://www.dynare.org/sebastien
  `-  GPG Key: 4096R/381A7594


pgp8fGrLmoMd7.pgp
Description: PGP signature

Re: major linux problems summary 2012

[Andrey Rahmatullin]

On Sun, Nov 04, 2012 at 12:41:13PM +0100, Michael Banck wrote:
> On Sun, Nov 04, 2012 at 04:01:03PM +0600, Andrey Rahmatullin wrote:
> > "You can't publish a (dynamically linked) binary that can be used on most
> > user systems"
> > "You cannot make all your GUI software look the same"
> 
> The information density and/or applicability of those statements is too
> low to usefully discuss them in the context of debian-devel, sorry.
debian-devel is not a proper level to discuss anything like that anyway. 

-- 
WBR, wRAR


signature.asc
Description: Digital signature

Re: Bug#692264: ITP: double-conversion -- routines to convert IEEE doubles to and from strings

[Roger Leigh]

On Sun, Nov 04, 2012 at 01:08:18PM +0100, Sébastien Villemot wrote:
> Roger Leigh  writes:
> 
> > On Sun, Nov 04, 2012 at 12:55:09PM +0100, Sébastien Villemot wrote:
> >> 
> >> * Package name: double-conversion
> >
> > libjs-double-conversion would be more appropriate, since
> > this is a javascript library.  double-conversion is too
> > generic.
> 
> It's actually written in C++, not JS, so I don't think the libjs prefix
> is appropriate.
> 
> The name "double-conversion" is taken from upstream, but I agree that it
> is quite generic. I don't have any better idea, but suggestions are
> welcome.

Sorry, misread the description there.  In that case, libdoubleconversion
would be better for the binary packages (though the source package name
doesn't need changing).


-- 
  .''`.  Roger Leigh
 : :' :  Debian GNU/Linuxhttp://people.debian.org/~rleigh/
 `. `'   schroot and sbuild  http://alioth.debian.org/projects/buildd-tools
   `-GPG Public Key  F33D 281D 470A B443 6756 147C 07B3 C8BC 4083 E800


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121104122410.gh14...@codelibre.net

Re: Bug#692264: ITP: double-conversion -- routines to convert IEEE doubles to and from strings

[Sébastien Villemot]

Roger Leigh  writes:

> On Sun, Nov 04, 2012 at 01:08:18PM +0100, Sébastien Villemot wrote:
>> Roger Leigh  writes:
>> 
>> > On Sun, Nov 04, 2012 at 12:55:09PM +0100, Sébastien Villemot wrote:
>> >> 
>> >> * Package name: double-conversion
>> >
>> > libjs-double-conversion would be more appropriate, since
>> > this is a javascript library.  double-conversion is too
>> > generic.
>> 
>> It's actually written in C++, not JS, so I don't think the libjs prefix
>> is appropriate.
>> 
>> The name "double-conversion" is taken from upstream, but I agree that it
>> is quite generic. I don't have any better idea, but suggestions are
>> welcome.
>
> Sorry, misread the description there.  In that case, libdoubleconversion
> would be better for the binary packages (though the source package name
> doesn't need changing).

Sure. I plan to use libdouble-conversion{,-dev,-dbg} (I prefer to keep
the hyphen in the binary package name, unless this is bad practice).

-- 
 .''`.Sébastien Villemot
: :' :Debian Developer
`. `' http://www.dynare.org/sebastien
  `-  GPG Key: 4096R/381A7594


pgpgfkYQk1qMU.pgp
Description: PGP signature

Re: IPv6, tentative addresses, bind(), wheezy

[Ben Hutchings]

On Sun, 2012-11-04 at 10:18 +0100, Marc Haber wrote:
> On Fri, 2 Nov 2012 23:24:55 +0100, martin f krafft
>  wrote:
> >The reason is that by the time bind() is called, the IPv6 address
> >(configured with /e/n/i inet6 static, which unbound should listen
> >on) is not yet ready, but "tentative", so the bind() call fails.
> 
> Last time when I asked a question along those lines (with bind as the
> application), the answer was the usual Debian-like "your application
> is broken, have upstream convert the app to an event-based approach
> and do not bother us".
> 
> This answer is partially right, applications need to be able to handle
> dynamically changing IP addresses at run-time to properly support
> IPv6. If they don't, one of the major advantages of IPv6 ("renumbering
> is easy") is lost. This goes especially if the app's configuration
> contains IP addresses.
> 
> The problem is, that the distributions, IMO, need to work around these
> shortcomings until the applications are eventually fixed (which might
> never happen). The socket API also has several shortcomings which
> makes this approach harder to do. For examle, there is no semantic for
> operations like "listen on this host address in all prefixes that the
> host system is aware of", probably written along the line of
> ::224:d7ff:fed0:5adc.

This wouldn't solve the problem that the UDP servers have, which is that
they need to be able to send replies from the same address the request
was sent to.  The old portable way to do this is to bind a socket to
each local address and reply using whichever socket the request was
received on.  That's why they enumerate local addresses rather than just
binding to 0.0.0.0 and ::.

For IPv6 the IPV6_PKTINFO socket option described in RFC 3542 allows
applications to get and set the local address per-packet; this is
hopefully implemented on most Unix-like systems.

For IPv4, Linux has a similar IP_PKTINFO option whereas BSD and Solaris
provide the options IP_{RECVDST,SENDSRC}ADDR.  This lack of portability
may explain why more applications have not adopted this approach yet.

Ben.

-- 
Ben Hutchings
No political challenge can be met by shopping. - George Monbiot


signature.asc
Description: This is a digitally signed message part

Re: IPv6, tentative addresses, bind(), wheezy

[martin f krafft]

also sprach Ben Hutchings  [2012.11.03.2027 +0100]:
> Whyever not?  You can get a socket bound to a non-local address
> even without this option, if the address is removed after you
> bind.  The restriction to current local addresses is only a sanity
> check which may or may not be useful.

You are right. ntpd does this nicely, but then again it's gonna take
us years to make all upstreams implement this properly. Might be
worth to figure out an intermediate solution too.

-- 
 .''`.   martin f. krafft   Related projects:
: :'  :  proud Debian developer   http://debiansystem.info
`. `'`   http://people.debian.org/~madduckhttp://vcs-pkg.org
  `-  Debian - when you have better things to do than fixing systems
 
"when women love us, they forgive us everything, even our crimes;
 when they do not love us, they give us credit for nothing,
 not even our virtues."
   -- honoré de balzac


digital_signature_gpg.asc
Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current)

Re: major linux problems summary 2012

[Russ Allbery]

Andrey Rahmatullin  writes:
> On Sat, Nov 03, 2012 at 02:37:34PM -0700, Russ Allbery wrote:

>> At least with open source software, a third path (besides working
>> around the bug or switching to some other software) also exists: you
>> can help.  That's not always practical, but at least it's *possible*.

> Not all problems are things that can be fixed by applying a text editor.
> Some of them are social, some would require major overhauls of major
> ecosystem components, probably in ways that will be rejected by current
> developers.

Sure.

My point is that this sort of thread will, if anything, make such changes
*less* likely to happen by causing people to feel obstinate and dig in
their heels.  The chances of it accomplishing anything productive are
remote.

So, if your goal is to fix these things, what you're doing right now is
directly counter-productive and will probably make the situation worse.
(It could be that you have some other goal than fixing things, such as
expressing frustration, or warning other users away from Linux due to bugs
that you think are serious, or trying to recruit new developers to do
something different, but I think most of the other possible goals are
off-topic on debian-devel.)

-- 
Russ Allbery (r...@debian.org)   


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87ehk9pbo5@windlord.stanford.edu

Re: IPv6, tentative addresses, bind(), wheezy

[martin f krafft]

also sprach Marc Haber  [2012.11.04.1013 +0100]:
> >Maybe for the time being it will be easier and safer to have ifupdown 
> >wait until DAD is finished?
> 
> How does one find out about that?

% ip -6 addr show dev eth0 | grep "2001:db8::deb1:46.* tentative"
inet6 2001:db8::deb1:46/64 scope global tentative

Once DAD completes, the "tentative" goes away.

Programatically, I think that's checking the return of
getifaddrs(3), ifa_flags, for IFA_F_TENTATIVE.

-- 
 .''`.   martin f. krafft   Related projects:
: :'  :  proud Debian developer   http://debiansystem.info
`. `'`   http://people.debian.org/~madduckhttp://vcs-pkg.org
  `-  Debian - when you have better things to do than fixing systems
 
an egg has the shortest sex-life of all: if gets laid once; it gets
eaten once. it also has to come in a box with 11 others, and the
only person who will sit on its face is its mother.


digital_signature_gpg.asc
Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current)

MBF: packages with autopkgtest tests but no XS-Testsuite

[Stefano Zacchiroli]

It looks like we have already 68 packages in the Debian archive shipping
autopkgtest [1] tests, which is great!  But unfortunately only 3 of them
declare the presence of the testsuite adding the recommended [2]
"XS-Testsuite: autopkgtest" header to their source stanza in
debian/control.

The dd-list of packages shipping autopkgtest tests but lacking the
header is attached. Please let me know if you notice any false positive.

I'd like to go ahead with a "Severity: wishlist" mass bug filing to
kindly ask maintainers to add the missing header.  The proposed mail
template for mass-bug is attached.

To avoid new instances of the problem in the future, I've also requested
a new lintian check [3]. In the meantime it'd be useful to ask
maintainers to add XS-Testsuite where needed.


Many thanks to Holger Levsen for setting up http://jenkins.debian.net/
and initiating this discussion on -qa [4], and to Jakub Wilk for making
me realize how many autopkgtest-enabled packages we already have in the
archive.

Cheers.

[1]: http://packages.qa.debian.org/a/autopkgtest.html
[2]: 
http://anonscm.debian.org/gitweb/?p=autopkgtest/autopkgtest.git;a=blob_plain;f=doc/README.package-tests;hb=HEAD
[3]: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=692282
[4]: https://lists.debian.org/debian-qa/2012/11/msg9.html
-- 
Stefano Zacchiroli  . . . . . . .  z...@upsilon.cc . . . . o . . . o . o
Maître de conférences . . . . . http://upsilon.cc/zack . . . o . . . o o
Debian Project Leader . . . . . . @zack on identi.ca . . o o o . . . o .
« the first rule of tautology club is the first rule of tautology club »
Adam Schmalhofer 
   apipkg
   execnet
   pytest-xdist

Andreas Tille 
   mafft (U)

Antonio Terceiro 
   pstreams
   rake (U)
   ruby-switch (U)

Arnaud Fontaine 
   python-mechanize (U)
   zc.buildout (U)
   zope.testbrowser (U)

Brian Sutherland 
   bobo (U)
   python-chameleon (U)
   python-mechanize (U)
   sourcecodegen (U)
   transaction (U)
   zc.buildout (U)
   zc.lockfile (U)
   zconfig (U)
   zdaemon (U)
   zodb (U)
   zope.authentication (U)
   zope.browser (U)
   zope.cachedescriptors (U)
   zope.component (U)
   zope.configuration (U)
   zope.contenttype (U)
   zope.copy (U)
   zope.dottedname (U)
   zope.event (U)
   zope.exceptions (U)
   zope.hookable (U)
   zope.i18n (U)
   zope.i18nmessageid (U)
   zope.interface (U)
   zope.location (U)
   zope.proxy (U)
   zope.publisher (U)
   zope.schema (U)
   zope.security (U)
   zope.sendmail (U)
   zope.sqlalchemy (U)
   zope.testbrowser (U)
   zope.testing (U)
   zope.testrunner (U)
   zope.traversing (U)

Charles Plessy 
   mafft (U)

Christian Perrier 
   samba4 (U)

Debian Bazaar Maintainers 
   bzr
   bzr-git
   bzr-rewrite
   bzr-stats
   bzr-svn
   bzr-upload

Debian Med Packaging Team 
   mafft

Debian Python Modules Team 
   gamera (U)
   python-byteplay (U)
   sphinx (U)

Debian QA Group 
   bzr-fastimport
   bzr-loom

Debian Ruby Extras Maintainers 

   rake
   ruby-switch

Debian/Ubuntu Zope Team 
   bobo
   python-chameleon
   python-mechanize
   sourcecodegen
   transaction
   zc.buildout
   zc.lockfile
   zconfig
   zdaemon
   zodb
   zope.authentication
   zope.browser
   zope.cachedescriptors
   zope.component
   zope.configuration
   zope.contenttype
   zope.copy
   zope.deprecation
   zope.dottedname
   zope.event
   zope.exceptions
   zope.hookable
   zope.i18n
   zope.i18nmessageid
   zope.interface
   zope.location
   zope.proxy
   zope.publisher
   zope.schema
   zope.security
   zope.sendmail
   zope.sqlalchemy
   zope.testbrowser
   zope.testing
   zope.testrunner
   zope.traversing

Fabio Tranchitella 
   bobo (U)
   python-chameleon (U)
   python-mechanize (U)
   sourcecodegen (U)
   transaction (U)
   zc.buildout (U)
   zc.lockfile (U)
   zconfig (U)
   zdaemon (U)
   zodb (U)
   zope.authentication (U)
   zope.browser (U)
   zope.cachedescriptors (U)
   zope.component (U)
   zope.configuration (U)
   zope.contenttype (U)
   zope.copy (U)
   zope.dottedname (U)
   zope.event (U)
   zope.exceptions (U)
   zope.hookable (U)
   zope.i18n (U)
   zope.i18nmessageid (U)
   zope.interface (U)
   zope.location (U)
   zope.proxy (U)
   zope.publisher (U)
   zope.schema (U)
   zope.security (U)
   zope.sqlalchemy (U)
   zope.testbrowser (U)
   zope.testing (U)
   zope.traversing (U)

Jakub Wilk 
   gamera
   ocrad
   python-byteplay
   sphinx

Jeff Bailey 
   bzr (U)

Jelmer Vernooij 
   bzr (U)
   bzr-git (U)
   bzr-rewrite (U)
   bzr-svn (U)
   python-fastimport
   samba4 (U)

Jérémy Bobbio 
   python-mechanize (U)

Koichi Akabe 
   bzr-stats (U)
   bzr-upload (U)

KURASHIKI Satoru 
   bzr-email

Ludovic Brenta 
   libgtkada

Maintainer: Debian QA Group 
   bzr-upload

martin f. krafft 
   pstreams (U)

Martin Pitt 
   udisks (U)

Matthias Klose 
   python-mechanize (U)
   zope.dottedname (U)
   zope.testing (U)

Michael Biebl 
   udisks (U)

Nicolas Boulenguez 
   libgmpada
   libgtkada (U)
   libncursesada
   libtexttools

Noèl Köthe 
   samba4 

Re: IPv6, tentative addresses, bind(), wheezy

[Steve Langasek]

On Sun, Nov 04, 2012 at 06:26:26PM +0100, martin f krafft wrote:
> also sprach Marc Haber  [2012.11.04.1013 +0100]:
> > >Maybe for the time being it will be easier and safer to have ifupdown 
> > >wait until DAD is finished?

> > How does one find out about that?

> % ip -6 addr show dev eth0 | grep "2001:db8::deb1:46.* tentative"
> inet6 2001:db8::deb1:46/64 scope global tentative

> Once DAD completes, the "tentative" goes away.

> Programatically, I think that's checking the return of
> getifaddrs(3), ifa_flags, for IFA_F_TENTATIVE.

Is there a way to get an event-based notification of this change, so
ifupdown can listen for such an event rather than having to poll?

-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
slanga...@ubuntu.com vor...@debian.org


signature.asc
Description: Digital signature

Re: MBF: packages with autopkgtest tests but no XS-Testsuite

[Paul Tagliamonte]

On Sun, Nov 04, 2012 at 07:01:12PM +0100, Stefano Zacchiroli wrote:
> It looks like we have already 68 packages in the Debian archive shipping
> autopkgtest [1] tests, which is great!  But unfortunately only 3 of them
> declare the presence of the testsuite adding the recommended [2]
> "XS-Testsuite: autopkgtest" header to their source stanza in
> debian/control.
> 
> The dd-list of packages shipping autopkgtest tests but lacking the
> header is attached. Please let me know if you notice any false positive.
> 
> I'd like to go ahead with a "Severity: wishlist" mass bug filing to
> kindly ask maintainers to add the missing header.  The proposed mail
> template for mass-bug is attached.

I think this sounds like a great idea, +1 to that.

> 
> To avoid new instances of the problem in the future, I've also requested
> a new lintian check [3]. In the meantime it'd be useful to ask
> maintainers to add XS-Testsuite where needed.
> 
> 
> Many thanks to Holger Levsen for setting up http://jenkins.debian.net/
> and initiating this discussion on -qa [4], and to Jakub Wilk for making
> me realize how many autopkgtest-enabled packages we already have in the
> archive.
> 
> Cheers.
> 
> [1]: http://packages.qa.debian.org/a/autopkgtest.html
> [2]: 
> http://anonscm.debian.org/gitweb/?p=autopkgtest/autopkgtest.git;a=blob_plain;f=doc/README.package-tests;hb=HEAD
> [3]: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=692282
> [4]: https://lists.debian.org/debian-qa/2012/11/msg9.html
> -- 
> Stefano Zacchiroli  . . . . . . .  z...@upsilon.cc . . . . o . . . o . o
> Maître de conférences . . . . . http://upsilon.cc/zack . . . o . . . o o
> Debian Project Leader . . . . . . @zack on identi.ca . . o o o . . . o .
> « the first rule of tautology club is the first rule of tautology club »


> [ This is an automated bug report, submitted as part of the mass bug
>   filing discussed at TODO-ADD-URL-TO-DEVEL-THREAD-HERE ]
> 
> According to the Contents file, #PACKAGE# ships a test suite for the
> autopkgtest testing framework [1]. Thanks a bunch for that, those test
> suites are a great way to improve the overall quality of the Debian
> archive!
> 
> However, the presence of test suites in packages should be easily
> discoverable, to allow test runners to pick suitable packages and run
> their test suites.  According to the autopkgtest specification [2], if
> you ship an autopkgtest test suite you should add the following header
> to the *source stanza* of your debian/control file:
> 
>   XS-Testsuite: autopkgtest
> 
> It seems that #PACKAGE# lacks such a header.
> Can you please add it?
> 
> Thanks for considering,
> Cheers.
> 
> 
> [1]: http://packages.qa.debian.org/a/autopkgtest.html
> [2]: 
> http://anonscm.debian.org/gitweb/?p=autopkgtest/autopkgtest.git;a=blob_plain;f=doc/README.package-tests;hb=HEAD

Content of the bug looks great to me too. Love the idea.

Cheers,
  Paul

-- 
 .''`.  Paul Tagliamonte 
: :'  : Proud Debian Developer
`. `'`  4096R / 8F04 9AD8 2C92 066C 7352  D28A 7B58 5B30 807C 2A87
 `- http://people.debian.org/~paultag


signature.asc
Description: Digital signature

Ayudeme por favor a iniciarme en el mundo de Linux

[Juan Perez]

Hola quiero aprender a manejar Linux, que distribución me recomendaría, Debian, 
Ubuntu, o Opensuse.
Me compre un libro de Linux, que era un plagio de otro libro de internet de 
otro autor, allí hablaba de CentOS, pero lei luego que CentOS, es mas para 
servidores.  ¿Ud que me recomendaría?
Leí que Ubuntu es Basura, yo no quiero decir nada, porque no se nada de Linux.
Descarge dos dvd de Linux que enseñan a utilizar Linux con fedora, y lei que 
fedora es el laboratorio de pruebas de red hat, y que no es recomendado para 
novatos como yo,
A mi me gustaría comenzar con debían, pero me dicen que hay que tener 
conocimientos para poder manejar debían, y yo no se nada de Linux. Seria bueno 
aprender Linux con Debian?
Leí que debían por ser tan libre, no es libre. ¿A que se refiere eso? ¿Es 
peligroso que la gente conozca internamente a Debían? ¿Afecta en algo a la 
seguridad de Debian? ¿Hace vunerable a  debían que la gente pueda ver su  
código?
¿Que distro de Linux me recomendaría para aprender Linux? ¿Me podría dar el 
link para descargar la distro de Linux en dvd? ¿Me recomendaría algún manual 
pdf de Linux, para descargarlo de internet? 
Leí que debían tiene dos niveles de ejecución por default,  red hat 3 o 5 
niveles de ejecución, ¿Qué es un nivel de ejecución o niveles de ejecución?
¿Tengo una duda cual es mejor libre office, o Open office o Microsoft Office?
Agradecere su pronta respuesta.
Gracias.  

Re: IPv6, tentative addresses, bind(), wheezy

[Cyril Brulebois]

Steve Langasek  (04/11/2012):
> Is there a way to get an event-based notification of this change, so
> ifupdown can listen for such an event rather than having to poll?

TL;DR: I think so.

I don't know anything about the linux kernel

Looking at net/ipv6/addrconf.c we have for examples:
  add_addr()
  addrconf_dad_start()

doing things like:
  ifp->flags &= ~IFA_F_TENTATIVE;
  ifp->flags &= ~(IFA_F_TENTATIVE|IFA_F_OPTIMISTIC|IFA_F_DADFAILED);

It looks like ipv6_ifa_notify() is called upon changes.

That one ends up calling inet6_ifa_notify(), which has things like
struct sk_buff, nlmsg_new(), rtnl_notify(). The last one is defined in
net/core/rtnetlink.c; it looks like somebody wants to hook oneself up
onto a netlink socket from userspace.

Mraw,
KiBi.


signature.asc
Description: Digital signature

Re: IPv6, tentative addresses, bind(), wheezy

[martin f krafft]

also sprach Steve Langasek  [2012.11.04.1921 +0100]:
> Is there a way to get an event-based notification of this change, so
> ifupdown can listen for such an event rather than having to poll?

I experimented with ip-monitor(8), which supposedly dumps RTNETLINK,
but it remained silent on the clearing of the tentative flag.

-- 
 .''`.   martin f. krafft   Related projects:
: :'  :  proud Debian developer   http://debiansystem.info
`. `'`   http://people.debian.org/~madduckhttp://vcs-pkg.org
  `-  Debian - when you have better things to do than fixing systems
 
perl -e 'print "The earth is a disk!\n" if ( "a" == "b" );'
   (dedicated to nori)


digital_signature_gpg.asc
Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current)

Re: IPv6, tentative addresses, bind(), wheezy

[Vincent Bernat]

 ❦  4 novembre 2012 19:21 CET, Steve Langasek  :

>> > How does one find out about that?
>
>> % ip -6 addr show dev eth0 | grep "2001:db8::deb1:46.* tentative"
>> inet6 2001:db8::deb1:46/64 scope global tentative
>
>> Once DAD completes, the "tentative" goes away.
>
>> Programatically, I think that's checking the return of
>> getifaddrs(3), ifa_flags, for IFA_F_TENTATIVE.
>
> Is there a way to get an event-based notification of this change, so
> ifupdown can listen for such an event rather than having to poll?

Trying with "ip monitor addr", it seems that the netlink notification is
sent once the tentative flags has been removed.
-- 
panic("Oh boy, that early out of memory?");
2.2.16 /usr/src/linux/arch/mips/mm/init.c


pgpKZoBfcLsoW.pgp
Description: PGP signature

Re: IPv6, tentative addresses, bind(), wheezy

[martin f krafft]

also sprach Vincent Bernat  [2012.11.04.2033 +0100]:
> Trying with "ip monitor addr", it seems that the netlink notification is
> sent once the tentative flags has been removed.

You are right. I have now removed the tomato from my eyes.

-- 
 .''`.   martin f. krafft   Related projects:
: :'  :  proud Debian developer   http://debiansystem.info
`. `'`   http://people.debian.org/~madduckhttp://vcs-pkg.org
  `-  Debian - when you have better things to do than fixing systems
 
logik ist analsadismus: gedanken werden gewaltsam
durch einen engen gang gepreßt.
-- frei nach lacan


digital_signature_gpg.asc
Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current)

Re: IPv6, tentative addresses, bind(), wheezy

[Ben Hutchings]

On Sun, 2012-11-04 at 19:48 +0100, Cyril Brulebois wrote:
> Steve Langasek  (04/11/2012):
> > Is there a way to get an event-based notification of this change, so
> > ifupdown can listen for such an event rather than having to poll?
> 
> TL;DR: I think so.
> 
> I don't know anything about the linux kernel

Lies!

> Looking at net/ipv6/addrconf.c we have for examples:
>   add_addr()
>   addrconf_dad_start()
> 
> doing things like:
>   ifp->flags &= ~IFA_F_TENTATIVE;
>   ifp->flags &= ~(IFA_F_TENTATIVE|IFA_F_OPTIMISTIC|IFA_F_DADFAILED);
> 
> It looks like ipv6_ifa_notify() is called upon changes.
> 
> That one ends up calling inet6_ifa_notify(), which has things like
> struct sk_buff, nlmsg_new(), rtnl_notify(). The last one is defined in
> net/core/rtnetlink.c; it looks like somebody wants to hook oneself up
> onto a netlink socket from userspace.

Right, you must use rtnetlink.  But that is no more portable than using
socket options to reflect addresses for UDP.

Ben.

-- 
Ben Hutchings
No political challenge can be met by shopping. - George Monbiot


signature.asc
Description: This is a digitally signed message part

Bug#692320: ITP: A tool for securing communications between a client and a DNS resolver

[sky blue]

Package: wnpp
Severity: wishlist
X-Debbugs-CC: debian-devel@lists.debian.org


Package name : dnscrypt-proxy
Version: 1.2.0
Upstream Author : Frank Denis 
URL  : http://dnscrypt.org
License: ISC
Description  : A tool for securing communications between
a client and a DNS resolver
dnscrypt-proxy provides local service which can be used directly as
your local resolver or as a DNS forwarder, encrypting and
authenticating requests using the DNSCrypt protocol and passing them
to an upstream server, by default OpenDNS who run this on their
resolvers.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CAMvSYRURgkuyq6mucPYtg33XzOj88Qg21=fcpawu7-k69xb...@mail.gmail.com

debian mate

[Виталий Короневич]

debian gnome-edition - ok
debian kde-edition - ok
debian lxde-edition - ok
debian xfce-edition - ok
when will debian mate-edition?

Re: debian mate

[Wouter Verhelst]

On Mon, Nov 05, 2012 at 02:57:35AM +0400, Виталий Короневич wrote:
> debian gnome-edition - ok
> debian kde-edition - ok
> debian lxde-edition - ok
> debian xfce-edition - ok
> when will debian mate-edition?

When someone (you?) does the work to package (the relevant parts of) mate.

-- 
Copyshops should do vouchers. So that next time some bureaucracy requires you
to mail a form in triplicate, you can mail it just once, add a voucher, and
save on postage.


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121104234140.gk18...@grep.be

Re: debian mate

[Andrew Kolotenko]

On Mon, 2012-11-05 at 02:57 +0400, Виталий Короневич wrote:
> debian gnome-edition - ok
> debian kde-edition - ok
> debian lxde-edition - ok
> debian xfce-edition - ok
> when will debian mate-edition?

what for? the ones above are enough imho


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1352072499.2362.2.camel@workstation.ground

Re: debian mate

[John Paul Adrian Glaubitz]

On Mon, Nov 05, 2012 at 02:57:35AM +0400, Виталий Короневич wrote:
> debian gnome-edition - ok
> debian kde-edition - ok
> debian lxde-edition - ok
> debian xfce-edition - ok

Those aren't really editions comparable to Ubuntu's editions or
Fedora's spins. You just install a regular Debian and choose the
desktop enviroment you want.

> when will debian mate-edition?

This isn't as easy and straight-forward as you might think. There are
many problems that need to be solved with MATE first before it can
enter Debian. Some of these are discussed here [1].

One of the biggest problems with MATE so far is that it currently
depends on libraries and other things that have been deprecated and
removed in Debian and should not be reintroduced to Debian. Thus, the
MATE developers are working to port MATE to more recent libraries
which are available in Debian.

Furthermore, work has to be done to make sure MATE co-installs
seamlessly with anything from GNOME which isn't that trivial. There
are many binaries and libraries which still exist in GNOME3 and
installing MATE could possibily break these.

I'm actually helping the MATE developers to get MATE ready for Debian,
but don't expect that to happen anytime soon.

I'm also working on getting mdm ready, a fork of gdm 2.20. Currently,
I'm not really happy with the code, however.

Adrian

> [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=658783


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121104235121.ga25...@physik.fu-berlin.de

Re: Bugs filed in unexpected places

[Tomas Pospisek]

Hello Andrei and all,

On Fri, 26 Oct 2012 16:24:59 +0300, Andrei POPESCU wrote:

> The discussion about ITO made me think: wouldn't it make more sense to 
> also have RFH, RFA, and O filled against the package itself and not 
> wnpp? One has to be quite familiar with Debian to check wnpp for RFH, 
> RFA or O. Maybe having these bugs "in the face" of people interested in 
> the package (i.e. on the package's bug page) can help attract 
> contributions.
> 
> Additionally for some packages it might make sense to remove them from 
> testing and raise the severity of the O bug to serious to signal that 
> the package should not be included in the next release unless someone is

> willing to commit to maintain it.
> 
> An immediate solution would probably be to 'affects ' so the 
> bugs at least shows up on the package's bug page. Maybe the BTS 
> could/should do this automatically?
> 
> 
> One a somewhat related note, I also notice confusion is created by the 
> removal bugs being filed against ftp.debian.org. When people not 
> familiar with Debian are looking into why a package has been removed 
> they look at the (archived) package's bugs. Not a biggie, but might help

> users or prospective ITPers (e.g. if the removal reasons still apply). 
> Not sure if 'affects' can help here.
> 
> I'm guessing the current procedures were created because at the time the

> BTS did not have the 'affects ' feature.

Have you tried to hack anything yet? I've seen that Don Armstrong seens to
be open to these ideas.

I've also been roaming around http://bugs.debian.org/bugs.debian.org and
I'm also encountering the same problems you describe above.
*t


-- 
To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/394759cd3e5b59c9f637b444f3df3...@sourcepole.ch