Re: db.debian.org (and related infrastructure) updates

[Andreas Tille]

On Wed, 3 Jan 2007, Kevin Mark wrote:


On Wed, Jan 03, 2007 at 09:31:26PM +0200, Debian Oracle wrote:

I hope that this explains everything, Kev. You owe the Oracle an e-mail
quotation trimming device.

...
more options. If its 'just' a field for our use, why does it need to use
a 'standard' that excludes. If there needs to be some interchange of
data in the future, we can certainly deal with this.


If I understand the great Debian Oracle right (BTW, for the moment the
funniest posting for this year on Debian lists - keep on the good work ;-) ),
every DD who is uncertain how to specify the own gender LDAP field is invited
to ask the Oracle for help which would be the right choice (for the moment
and in future).

Unfortunately I fail to see in how far the definition of a gender field
in db.debian.org would bring us closer or farer to our goal to release
the best operating system.  So I would like to suggest to move this
thread to debian-curiosa.

Happy new year to all list members

 Andreas.

--
http://fam-tille.de


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: etch's upgrades during life cycle

[Luis Matos]

Qua, 2007-01-03 às 22:28 +0100, Daniel Baumann escreveu:
> Luis Matos wrote:
> > So, if we loose security and stability ... why use debian?
> 
> security and stability, that is excately what makes these backports
> unofficial (more stability and bugs are an issue than security, though).
> 
> however, if you want to have latest and greatest but with stability and
> security as you know it from debian stable, then you are asking for the
> impossible.

i don't want the latest and the greatest ... i want a newer kernel :P
not means to be the latest ... even when you know that there are some
buggy releases.

> 
> -- 
> Address:Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
> Email:  [EMAIL PROTECTED]
> Internet:   http://people.panthera-systems.net/~daniel-baumann/
> 
> 
-- 
Best Regards,
--
Luis Matos


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: etch's upgrades during life cycle

[Moritz Muehlenhoff]

Luis Matos wrote:
> Many users have complaints about in the middle of the life cycle, or
> before the debian stable release no longer supports new hardware.
> Therefor a new kernel would be needed for d-i ( or an hardware
> compatibility update for the kernel and modules).
>
> My proposal would be in point releases to change the kernel a bit to
> support more hardware. That kernel would be tested, ofcourse.
>
> What i am saying is: is it possible to in a lenny or lenny++ change the
> way debian upgrades it's stable, just for the kernel?

Yes, there are plans for a second set of kernels (and probably xservers)
nine months after Etch release, which will also have security support.

However nothing's fixed yet, as the current focus is on getting Etch ready.

Cheers,
Moritz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Some of us are worrying

[Geneva Barrett]

<>

Re: RFC: Proposal for official screenshot repo

[Pierre Habouzit]

On Thu, Jan 04, 2007 at 12:05:03AM +0100, Nico Golde wrote:
> Hi,
> * Pierre Habouzit <[EMAIL PROTECTED]> [2007-01-03 23:53]:
> > On Wed, Jan 03, 2007 at 05:32:52PM -0500, Roberto C. Sanchez wrote:
> > > Today I was trying to explain to a friend the concept of using a package
> > > management front end to search for and install packages.  He liked the
> > > idea of descriptions, but found it hard to imagine what some programs
> > > looked like.  It occurs to me that what is missing is screenshots.
> > 
> >   this idea has been discussed recently, I just can't remember when
> > exactly though, I'd say in the late 6 monthes. Maybe you can grab some
> > names of people that were involved in the first proposal there. It may
> > have been proposed on -project rather than devel. But I suppose google
> > will know about it.
> 
> Maybe the following?
> http://lists.debian.org/debian-devel/2006/04/msg00915.html
> Some replies are not in the thread.
> Kind regards
> Nico

  yes, that was it, thanks. the following is on the next month archive,
hence why those are missing :)

  http://lists.debian.org/debian-devel/2006/05/

  that was older that what I thought btw :)
-- 
·O·  Pierre Habouzit
··O[EMAIL PROTECTED]
OOOhttp://www.madism.org


pgpKSzgsVphxF.pgp
Description: PGP signature

Re: RFC: Proposal for official screenshot repo

[Nico Golde]

Hi,
* Pierre Habouzit <[EMAIL PROTECTED]> [2007-01-03 23:53]:
> On Wed, Jan 03, 2007 at 05:32:52PM -0500, Roberto C. Sanchez wrote:
> > Today I was trying to explain to a friend the concept of using a package
> > management front end to search for and install packages.  He liked the
> > idea of descriptions, but found it hard to imagine what some programs
> > looked like.  It occurs to me that what is missing is screenshots.
> 
>   this idea has been discussed recently, I just can't remember when
> exactly though, I'd say in the late 6 monthes. Maybe you can grab some
> names of people that were involved in the first proposal there. It may
> have been proposed on -project rather than devel. But I suppose google
> will know about it.

Maybe the following?
http://lists.debian.org/debian-devel/2006/04/msg00915.html
Some replies are not in the thread.
Kind regards
Nico

-- 
Nico Golde - http://www.ngolde.de
JAB: [EMAIL PROTECTED] - GPG: 0x73647CFF
Forget about that mouse with 3/4/5 buttons,
gimme a keyboard with 103/104/105 keys!


pgpJ7RWGAYkbD.pgp
Description: PGP signature

Re: RFC: Proposal for official screenshot repo

[Pierre Habouzit]

On Wed, Jan 03, 2007 at 05:32:52PM -0500, Roberto C. Sanchez wrote:
> Today I was trying to explain to a friend the concept of using a package
> management front end to search for and install packages.  He liked the
> idea of descriptions, but found it hard to imagine what some programs
> looked like.  It occurs to me that what is missing is screenshots.

  this idea has been discussed recently, I just can't remember when
exactly though, I'd say in the late 6 monthes. Maybe you can grab some
names of people that were involved in the first proposal there. It may
have been proposed on -project rather than devel. But I suppose google
will know about it.
-- 
·O·  Pierre Habouzit
··O[EMAIL PROTECTED]
OOOhttp://www.madism.org


pgpmMZTPILzeF.pgp
Description: PGP signature

What happened to Agnula.org (DeMuDi)?

[RalfGesellensetter]

Did the domain expire?

demudi.org is full of dead links now.
 
[EMAIL PROTECTED] is dead, too.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

RFC: Proposal for official screenshot repo

[Roberto C. Sanchez]

Today I was trying to explain to a friend the concept of using a package
management front end to search for and install packages.  He liked the
idea of descriptions, but found it hard to imagine what some programs
looked like.  It occurs to me that what is missing is screenshots.

Please note that this idea is targeted at GUI users, particularly those
who also use a GUI package manager (e.g., synaptic), though not
exclusively.

What I propose is this:

A service located at screenshots.debian.org (or something else
appropriate), similar to packages.debian.org.  Naturally, every package
will not require screenshots, but for those which would benefit from
letting the user see them in action, this could work.

It would be necessary to add a field in debian/control called
"Screenshots:" which provides a URL to the screenshots.  The package
manager can handle this in different ways.  For example, aptitude may
simply want to print the URL so that if the user is using it in an
xterm, he can copy and paste into a browser.  OTOH, synaptic might
simply create a button (without showing the URL) that launches a browser
to the right window.  Alternatively, synaptic may download the images
and just display them in a tab or part of the window.  The possibilties
are many.

Users could also simply browse to screenshots.d.o and search and browse
as with packages.d.o.

Uploads of screenshots can be handled similarly to uploads now, either
by a recognized DD or a DD sponsoring someone else.  Another possibility
would be for something which requires a web login and allows direct
uploading.  We would need a way to tag the screenshots with version
numbers and other suitable metadata to avert confusion for users.
Additionally, we may want to, as p.d.o does, differentiate between
oldstable/stable/testing/experimental.  User contributed screenshots can
be submitted by filing wishlist bugs against the package and the
maintainer can then choose whether or not to accept the contribution.

There would need to be standards for size and format.  I would propose
640x480 or 480x640 (soft minimum) to 1024x768 or 768x1024 (hard
maximum), PNG with max compression.  Max file size could be set at 100k
or something like that.

I'd be interested in knowing:

 * Would such an idea be feasible?
 * Would maintainers be willing to occasionally upload screenshots?
 * What would be a good way to get started? (*)

As far as getting started:

 * Would this need to start on debian.net?
 * What would be the requirements/limitations/guidelines?

Did I miss anything?

I would really appreciate any comments and suggestions on this.

Regards,

-Roberto

-- 
Roberto C. Sanchez
http://people.connexer.com/~roberto
http://www.connexer.com


signature.asc
Description: Digital signature

Re: etch's upgrades during life cycle

[Daniel Baumann]

Luis Matos wrote:
> So, if we loose security and stability ... why use debian?

security and stability, that is excately what makes these backports
unofficial (more stability and bugs are an issue than security, though).

however, if you want to have latest and greatest but with stability and
security as you know it from debian stable, then you are asking for the
impossible.

-- 
Address:Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
Email:  [EMAIL PROTECTED]
Internet:   http://people.panthera-systems.net/~daniel-baumann/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: etch's upgrades during life cycle

[Luis Matos]

Qua, 2007-01-03 às 22:13 +0100, Daniel Baumann escreveu:
> Luis Matos wrote:
> > What i am saying is: is it possible to in a lenny or lenny++ change the
> > way debian upgrades it's stable, just for the kernel?
> 
> both things are already solved unofficially. there are kernel backports
> [0], and kenshi makes stable-with-new-kernel installer-images[1].
> 
> so, basically, you are asking to make them (more) official now?

basically yes.
When a user approaches debian, goes to www.debian.org then follows the
link to the distribution, grabs the iso, burns it and install it.

when it does not fit ... he will say that debian does not support
hardware foo (this can be truth) and goes for another distribution.

Even, i don't know if the d-i images posted there have security updates.
People know that a kernel that is debian supported is secure and bug
free (or more bug free than others). So, if we loose security and
stability ... why use debian?

> 
> [0] http://www.backports.org/debian/pool/main/l/linux-2.6/
> [1] http://kmuto.jp/debian/d-i/
> 
> -- 
> Address:Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
> Email:  [EMAIL PROTECTED]
> Internet:   http://people.panthera-systems.net/~daniel-baumann/
> 
> 
-- 
Best Regards,
--
Luis Matos


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: db.debian.org (and related infrastructure) updates

[Kevin Mark]

On Wed, Jan 03, 2007 at 09:31:26PM +0200, Debian Oracle wrote:
> I hope that this explains everything, Kev. You owe the Oracle an e-mail
> quotation trimming device.
Greetings O great Oracle, I did manage to extract most of the meaning
out of the consise phrases electronically transmitted by the mystic
vorlon into the great internet tubes.  I just have expectations that if
there is going to be progress in the recognition of the multitude of
human natures by which people define themselves, that the 'Free'
software world would be at the forefront of that effort. ISO is a
'standard'. Google shows a few threads about the inadequet nature of
this standard. I just sent an email about a DICOM document that includes
more options. If its 'just' a field for our use, why does it need to use
a 'standard' that excludes. If there needs to be some interchange of
data in the future, we can certainly deal with this.
cheers,
Kev
-- 
|  .''`.  == Debian GNU/Linux == |   my web site:   |
| : :' :  The  Universal |   'under construction'   |
| `. `'  Operating System| go to counter.li.org and |
|   `-http://www.debian.org/ |be counted! #238656   |
| my keysever: pgp.mit.edu   | my NPO: cfsg.org |


signature.asc
Description: Digital signature

Re: etch's upgrades during life cycle

[Daniel Baumann]

Luis Matos wrote:
> What i am saying is: is it possible to in a lenny or lenny++ change the
> way debian upgrades it's stable, just for the kernel?

both things are already solved unofficially. there are kernel backports
[0], and kenshi makes stable-with-new-kernel installer-images[1].

so, basically, you are asking to make them (more) official now?

[0] http://www.backports.org/debian/pool/main/l/linux-2.6/
[1] http://kmuto.jp/debian/d-i/

-- 
Address:Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
Email:  [EMAIL PROTECTED]
Internet:   http://people.panthera-systems.net/~daniel-baumann/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: db.debian.org (and related infrastructure) updates

[Kevin Mark]

On Wed, Jan 03, 2007 at 01:47:48PM -0500, Kevin Mark wrote:
> On Tue, Jan 02, 2007 at 06:32:10PM -0800, Steve Langasek wrote:
> > On Wed, Jan 03, 2007 at 12:50:27AM +0100, Wouter Verhelst wrote:
> > > [EMAIL PROTECTED]:~$ ldapsearch -h 'db.debian.org' -b'cn=Subschema' -x -s 
> > > base '(objectClass=*)' attributeTypes | grep gender
> > > attributeTypes: ( 1.3.6.1.4.1.9586.100.4.2.30 NAME 'gender' DESC 'ISO 
> > > 5218 rep
> > >  resentation of human gender' EQUALITY integerMatch SYNTAX 
> > > 1.3.6.1.4.1.1466.11
> > 
> > > In other words, if you want to see that changed, take it up with ISO.
> > > No, changing it unilaterally in Debian won't help, either; In LDAP, a
> > > field of a specific name always (*always*) has a certain syntax;
> > > breaking that would break software that expects this particular field to
> > > be in that particular syntax.
> > 
> > That's not completely true; you could have an attribute type named 'gender'
> > with a different OID and different syntax/semantics, you just wouldn't be
> > able to use it on an LDAP server which also needed the use of the ISO
> > attribute type or of any object classes that are defined to use the ISO
> > attribute.
> > 
> > But if all of our Japanese, Chinese, Greek Orthodox, Muslim, and French
> > Revolutionary developers can tolerate having to enter their birthdates using
> > the Gregorian calendar, I think we'll be able to make do with an opt-in
> > binary gender classification too.
> Hi Steve,
> I have yet to see a use case for this LDAP item. Is it strictly for a
> male/femaie survey that other FLOSS projects will join? Does this mean
> that people who dont self-identify as male or female are just not
> counted? According to some stats that could be 100 people.  Is there any
> ISO standard that is inclusive of those uncounted people?
> cheers,
I found a document for DICOM that includes more options
cheers,
Kev
[0] http://medical.nema.org/Dicom/CP/CPack_23/cp373_lb.pdf
-- 
|  .''`.  == Debian GNU/Linux == |   my web site:   |
| : :' :  The  Universal |   'under construction'   |
| `. `'  Operating System| go to counter.li.org and |
|   `-http://www.debian.org/ |be counted! #238656   |
| my keysever: pgp.mit.edu   | my NPO: cfsg.org |


signature.asc
Description: Digital signature

Re: db.debian.org (and related infrastructure) updates

[Debian Oracle]

On ke, 2007-01-03 at 13:47 -0500, Kevin Mark wrote:
> I have yet to see a use case for this LDAP item. Is it strictly for a
> male/femaie survey that other FLOSS projects will join? Does this mean
> that people who dont self-identify as male or female are just not
> counted? According to some stats that could be 100 people.  Is there any
> ISO standard that is inclusive of those uncounted people?

You have reached the Debian Oracle. Please allow the Oracle to translate
Steve's message to plain English. Steve is a great guy, but he
occasionally uses difficult words and constructs of grammar, and those
can sometimes confuse the rest of us. He is the victim of a childhood
spent in a Catholic orphanage run by Latin-speaking priests, so he grew
up thinking "alea iacta est" was a normal way of saying "yes, sir, I
will fix a release critical bug at once, sir, thank you sir".

The key phrase in Steve's verbiage is "I think we'll be able to make do
with an opt-in binary gender classification". 

"I think" is a pair of words that is often used to indicate personal
opinion, so Steve uses it to say that what he says next is what the
project should do as far as he is concerned, but that it isn't official
Debian policy.

"Make do" is another important word pair, which means "manage to suffer
without excessive or undue pain". Here Steve indicates that although the
solution chosen is not perfect, it is good enough at least for now, and
gives the implication that we have more important things to worry about.

The third really significant part is "opt-in binary gender
classification".

"Binary gender classification" is Steve's Latinesque way of saying
"there are two genders to choose from". In this case, there's two
choices; by implication, they are "male" and "female" rather than "C"
and "C++".

With "opt-in" Steve means that Debian developers may opt, er, in, into
telling everyone whether they're male or female. That means they can do
it if they want to, or not do it if they don't want to. In some cases,
if the other available choices are inappropriate for them, the might not
be able to fill it in, but "opt-in" covers that, too. So those who want
to, and are able to, to choose from the two gender options can do so,
and everyone else can choose neither. So actually there are three
values: male, female, and unspecified.

This should cover the central part of your message: people who do not
identify themselves as "male" or "female" can choose "unspecified".

>From a vast experience in dealing with humankind, the Debian Oracle
further provides the following statements to further respond to your
question: The use case for this field is purely statistical, but it is
in no way tied to any existing or planned FLOSS surveys or other
projects than Debian. The ISO does not have a non-binary gender
classification system that Debian could use. If we want to make the
statistics classify every person's gender exactly, the field needs to be
free-form text.

I hope that this explains everything, Kev. You owe the Oracle an e-mail
quotation trimming device.



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

etch's upgrades during life cycle

[Luis Matos]

Hello

i m a debian user and like many other debian users there are things in
debian that i like and dislike.

I am going to get a round no for what i am asking, but i think it is a
good question.

Many users have complaints about in the middle of the life cycle, or
before the debian stable release no longer supports new hardware.
Therefor a new kernel would be needed for d-i ( or an hardware
compatibility update for the kernel and modules).

My proposal would be in point releases to change the kernel a bit to
support more hardware. That kernel would be tested, ofcourse.

What i am saying is: is it possible to in a lenny or lenny++ change the
way debian upgrades it's stable, just for the kernel?

Other programs can be upgraded by volatile's repository or backports but
the kernel is something that is crucial. We (users) just want to enter
the cd in the cdrom reader and ... get debian installed in 15 minutes.

i know the work that would be needed and probably the bugs that would
come up... another choice would be a new package of modules, with the
correspondent udeb for d-i.

Please, don't kill me because someone has to talk about this once in a
while.


-- 
Best Regards,
--
Luis Matos


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: db.debian.org (and related infrastructure) updates

[Kevin Mark]

On Tue, Jan 02, 2007 at 06:32:10PM -0800, Steve Langasek wrote:
> On Wed, Jan 03, 2007 at 12:50:27AM +0100, Wouter Verhelst wrote:
> > [EMAIL PROTECTED]:~$ ldapsearch -h 'db.debian.org' -b'cn=Subschema' -x -s 
> > base '(objectClass=*)' attributeTypes | grep gender
> > attributeTypes: ( 1.3.6.1.4.1.9586.100.4.2.30 NAME 'gender' DESC 'ISO 5218 
> > rep
> >  resentation of human gender' EQUALITY integerMatch SYNTAX 
> > 1.3.6.1.4.1.1466.11
> 
> > In other words, if you want to see that changed, take it up with ISO.
> > No, changing it unilaterally in Debian won't help, either; In LDAP, a
> > field of a specific name always (*always*) has a certain syntax;
> > breaking that would break software that expects this particular field to
> > be in that particular syntax.
> 
> That's not completely true; you could have an attribute type named 'gender'
> with a different OID and different syntax/semantics, you just wouldn't be
> able to use it on an LDAP server which also needed the use of the ISO
> attribute type or of any object classes that are defined to use the ISO
> attribute.
> 
> But if all of our Japanese, Chinese, Greek Orthodox, Muslim, and French
> Revolutionary developers can tolerate having to enter their birthdates using
> the Gregorian calendar, I think we'll be able to make do with an opt-in
> binary gender classification too.
Hi Steve,
I have yet to see a use case for this LDAP item. Is it strictly for a
male/femaie survey that other FLOSS projects will join? Does this mean
that people who dont self-identify as male or female are just not
counted? According to some stats that could be 100 people.  Is there any
ISO standard that is inclusive of those uncounted people?
cheers,
Kev
-- 
|  .''`.  == Debian GNU/Linux == |   my web site:   |
| : :' :  The  Universal |   'under construction'   |
| `. `'  Operating System| go to counter.li.org and |
|   `-http://www.debian.org/ |be counted! #238656   |
| my keysever: pgp.mit.edu   | my NPO: cfsg.org |


signature.asc
Description: Digital signature

Re: FSG Packaging Summit in Berlin

[Sebastian Feltel]

Hello,

Ottavio Caruso schrieb am 03.01.2007 15:35:
> http://www.freestandards.org/en/LSB_face-to-face_%28December_2006%29
> 
> Quote:"The goal of the Summit is to bring together the key people in
> the Linux packaging world and ISVs to discuss the future of Linux
> packaging. Topics will include RPM, the role of other packaging
> technologies (dpkg, APT, yum, etc.), and the needs of ISVs in
> packaging solutions".
> 
> Has this ever been discussed on Debian ML's?
> 
> Joey Hess was there. I can't recall any mention of it on the DWN.

If you are aware of currently unmentioned events, news or whatever
please report [1] them. We (meaning the regular contributors to DWN)
cannot read every single Debian/Linux/FOSS-related mailing list or
website so we need your support.

Bye
Sebastian

[1] http://www.debian.org/News/weekly/contributing

-- 
debianforum.de - die deutschsprachige Supportwebseite rund
um das Debian-Projekt  



signature.asc
Description: OpenPGP digital signature

Re: Kernel 2.6.17 and 2.6.18-3-K7 network problem [SOLVED]

[Antonio Laterza]

--- Antonio Laterza <[EMAIL PROTECTED]> ha scritto:

> Hi all, I posted the question below to debian-user ...
> maybe nobody experienced my same issue. How can I
> contribute to eventually solve? Which other information
> are
> necessary in order to focalize the problem?
> 
> 
> I want to ask if somebody experienced the same
> problem I have with etch installation and in general with
> kernel 2.6.17 and 2.6.18 on AMD K7. The computer where I
> tried has 2 network cards, I connect only 1 card and it
> receives the IP address.

Hi all, just to report the solution was suggested by
Hendrik Sattler on debian-devel list.
> TCP windows scaling comes to mind:
> http://lwn.net/Articles/92727/
> 
> HS
Wow!!! Best regards Hendrik!!!
Most probabily the provider's router is buggy but it solved
the problem.
The solution was that reported by article pointed by you:

In the mean time, anybody running a current kernel who is
having trouble connecting to a needed site can work around
the problem with a command like:

echo 0 > /proc/sys/net/ipv4/tcp_default_win_scale 

or by adding a line like:

net.ipv4.tcp_default_win_scale = 0

to /etc/sysctl.conf.



anlater
http://studioinglaterza.blogspot.com/


__
Do You Yahoo!?
Poco spazio e tanto spam? Yahoo! Mail ti protegge dallo spam e ti da tanto 
spazio gratuito per i tuoi file e i messaggi 
http://mail.yahoo.it 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Kernel 2.6.17 and 2.6.18-3-K7 network problem [SOLVED]

[Antonio Laterza]

--- Hendrik Sattler <[EMAIL PROTECTED]> ha scritto:

> Am Mittwoch 03 Januar 2007 09:53 schrieb Antonio Laterza:
> > :( ). If I choose the kernel 2.6.18 X is OK but I am
> not
> >
> > able to reache many sites. I can navigate very few
> sites
> > example: www.google.com and www.mozilla.org. Using
> ethereal
> > seems that the HTTP request go out (i.e. GET / )
> but
> > the ACK packet does not ask the next packet number 
> in
> > other words seems that the GET request never reched the
> > final site.
> 
> TCP windows scaling comes to mind:
> http://lwn.net/Articles/92727/
> 
> HS
Wow!!! Best regards Hendrik!!!
Most probabily the provider's router is buggy but it solved
the problem.
The solution was that reported by article pointed by you:

In the mean time, anybody running a current kernel who is
having trouble connecting to a needed site can work around
the problem with a command like:

echo 0 > /proc/sys/net/ipv4/tcp_default_win_scale 

or by adding a line like:

net.ipv4.tcp_default_win_scale = 0

to /etc/sysctl.conf.



anlater
http://studioinglaterza.blogspot.com/


__
Do You Yahoo!?
Poco spazio e tanto spam? Yahoo! Mail ti protegge dallo spam e ti da tanto 
spazio gratuito per i tuoi file e i messaggi 
http://mail.yahoo.it 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: FSG Packaging Summit in Berlin

[Kevin Mark]

On Wed, Jan 03, 2007 at 06:35:02AM -0800, Ottavio Caruso wrote:
> http://www.freestandards.org/en/LSB_face-to-face_%28December_2006%29
> 
> Quote:"The goal of the Summit is to bring together the key people in
> the Linux packaging world and ISVs to discuss the future of Linux
> packaging. Topics will include RPM, the role of other packaging
> technologies (dpkg, APT, yum, etc.), and the needs of ISVs in
> packaging solutions".
> 
> Has this ever been discussed on Debian ML's?
> 
> Joey Hess was there. I can't recall any mention of it on the DWN.
> 
> Why do we [users] have to learn of these things from external
> sources?
> 
> Regards
> 
> -- 
Hi Ottavio,
are you suggesting that Joey(or any DD) should have announced his/her
involvment in this type of event on a Debian mailing list? 
Maybe the event was not open to the public?
Is there a mailing list where DD's mention events they are attending on
behalf of Debian? Or is there a mailing list of events of interest to
Debian and its users. The only one that seem close is :
debian-events-{eu,us,be}.
cheers,
Kev
-- 
|  .''`.  == Debian GNU/Linux == |   my web site:   |
| : :' :  The  Universal |   'under construction'   |
| `. `'  Operating System| go to counter.li.org and |
|   `-http://www.debian.org/ |be counted! #238656   |
| my keysever: pgp.mit.edu   | my NPO: cfsg.org |


signature.asc
Description: Digital signature

Re: db.debian.org (and related infrastructure) updates

[Adrian von Bidder]

On Monday 01 January 2007 22:20, Josselin Mouette wrote:
> Le lundi 01 janvier 2007 à 17:51 +0100, Marco d'Itri a écrit :
> > On Jan 01, Josselin Mouette <[EMAIL PROTECTED]> wrote:
> > > rejecting email blindly based on data as
> > > reliable as RBLs is likely to give tons of false positives.
> >
> > This can be easily disproven by anybody who does this...
>
> Of course. I'm pretty sure that nobody on this list has ever got emails
> rejected because of broken RBLs.

And of course having one or two mails (that I can remember) rejected because 
of borked RBLs is "tons of false positives"?

Besides: Linux has tons of bugs.  It still solves many of my computing 
problems.  RBLs are probably not the golden bullet either, but they're an 
important part of my spam prevention measures, and I could even remove 
the "send spam (as per spamassassin) to [EMAIL PROTECTED] to devnull" hack, 
which is much more prone to false positives, and where the false positives 
are much, much worse (senders get no indication at all) than with RBLs, 
where the sender get a bounce.

Greylisting and callout verfication are to other pieces in the puzzle, the 
latter being the one I find the most controversial, the first one being the 
one that spammers are slowly getting the hang of.  (But if the RBL get fast 
enough so that a spam sender is in the RBL by the time the sender tries to 
send the spam the 2nd time, I still have won :-)

All of these are much, much more preferrable to all measures that can only 
be used when the mail body is on my server, because (i) sending mailservers 
often don't deal properly with rejections at the DATA stage and (ii) if 
rejection is not an option, and dropping is IMHO not a good option either, 
I'll still have to look through my spam folder.

cheers
-- vbi

-- 
Shutting down networkservers reguarly during worktime prevents RSI and
develops social contacts at work.


pgpAQ6YrZupH4.pgp
Description: PGP signature

Re: db.debian.org (and related infrastructure) updates

[Matthias Julius]

Santiago Vila <[EMAIL PROTECTED]> writes:

> If your SMTP server is listed in a DNSBL which I told db.debian.org
> to use for my debian.org email and you try to send me a message,
> then master will say "I don't accept this message" to your SMTP
> server, and your SMTP server, in turn, will send you the usual
> mailer-daemon message saying "Undelivered Mail Returned to Sender".

This sounds much better.  I was just thinking of occasional emails I
get saying: "Your email sent to 
was classified as spam. ..."

>
> I was comparing the previous scenario with the current one. The risk
> of missing an email because of it being lost inside a very big spam
> folder is now very low. This is one of the reasons rejecting a lot of
> email at SMTP time and filtering the rest (what we can do now) is
> usually better than not rejecting anything at all and trying to "filter"
> everything afterwards.

I agree.

Matthias


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: FSG Packaging Summit in Berlin

[Daniel Baumann]

Ottavio Caruso wrote:
> Why do we [users] have to learn of these things from external
> sources?

debian is volunteer project[0].

maybe you [user] should support debian when it does not fulfil your
expectations. if you can not contribute[1] personally, also donations[2]
are welcome.

[0] http://www.debian.org/devel/constitution
[1] http://www.debian.org/devel/join
[2] http://www.debian.org/donations

-- 
Address:Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
Email:  [EMAIL PROTECTED]
Internet:   http://people.panthera-systems.net/~daniel-baumann/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Announce: zerotools (tools to keep virtual machine disks cleaner)

[Aleksandr Koltsoff]

Hello,
Amaya wrote:
> I maintain the package perforate. It contains a very similar utility,
> zum, that is just a small c binary that seems to do the same.

Thanks for the perforate info.

zum is a resparser for existing files. This is quite different from
filling existing files with binary zero. A resparsing done inside
virtual disks would not make things any easier when compressing the
virtual disk (from outside the vm).

Many people seem to misunderstand the exact problem which zerotools
tries to solve though :-).

> Also take a look at http://bugs.debian.org/294297 and be warned that
> these kind of utilities can ruin a vfat file system (just warning you
> in case you have not thought about it).

One could probably implement the resparsing operation that zum does
using a similar LD_PRELOAD technique, but as you note yourself, sparse
files have interesting properties when the underlying filesystem does
not support sparse files. Also creating new files from inside unlink()
for doing resparsing doesn't at all sound like a thing that I'd like to
see in my systems. zerounlink never creates new files, hence it doesn't
trip over full inode tables and such.

Summa summarum, I don't think that zerotools and perforate/zum overlap
at all. The closest parallel between zum and zerotools is in zerofile,
which is just a mindless utility. I don't imagine people will use
zerofile a lot since the LD_PRELOAD-mechanism (with the wrapper) is much
easier and automatic.

ak.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Announce: zerotools (tools to keep virtual machine disks cleaner)

[Aleksandr Koltsoff]

Hi again,

Eduard Bloch wrote:
> Please add a technical overview to your introduction. As a techie, I
> would like to read what is so special about zerotools what I cannot do
> with:
> rm crapfiles && cat /dev/zero > bigfile && sync && rm bigfile && sync.
> 
> And you introduction does not tell me about anything new in your tools
> and I do not like reading a long novel to learn that simple details.

Thank you for the suggestion. I've modified the intro to cover the first
case, as for the second case (comparison to "bigfile"-technique), that
is already covered in "Other solutions". The "bigfile"-technique is
inferior when one uses virtual disks which are "sparse" (or dynamic as
some virtualization software calls them). When using "flat" (or
preallocated) virtual disk files, the bigfile technique is not so bad,
but personally I prefer an automatic mechanism so that I don't need to
do manual filling or take the virtual system to runlevel 1 specifically
for "bigfile".

ak.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

FOTOGRAFIA: cursos de verano

[aprende]

Hola!..Respondiendo a tu inquietud, te cuento que nosotros en este momento  no 
estamos dando talleres de fotografia durante el verano, pero en La Escuela de 
fotografia  MOTIVARTE, (Buenos Aires)  se estan dando cursos intensivos de 
verano de 1 mes de duracion, que son muy recomendables,  consulta con ellos. Te 
paso la web www.motivarte.com 
Cordiales Saludos
Ana




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#405429: ITP: cortado -- streaming applet for Ogg formats

[Torsten Werner]

Package: wnpp
Severity: wishlist
Owner: Torsten Werner <[EMAIL PROTECTED]>

* Package name: cortado
  Version : 0.2.2
  Upstream Authors: Wim Taymans, J^, Thomas Vander Stichele
* URL : http://www.flumotion.net/cortado/
* License : GPL and LGPL
  Programming Lang: Java
  Description : streaming applet for Ogg formats

 By embedding cortado applet in your website, you can give viewers access to
 streams from either the Flumotion streaming server or play a local
 file from your server without the need for a locally installed media
 player supporting the correct formats on the visitori's computer.
 .
 Cortado currently include Java decoders for Ogg Theora, Ogg Vorbis,
 Mulaw audio, MJPEG and our own Smoke codec.
 .
  Homepage: http://www.flumotion.net/cortado/


svn-buildpackage stuff can be found at
http://bollin.googlecode.com/svn/cortado/trunk/ .



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

FSG Packaging Summit in Berlin

[Ottavio Caruso]

http://www.freestandards.org/en/LSB_face-to-face_%28December_2006%29

Quote:"The goal of the Summit is to bring together the key people in
the Linux packaging world and ISVs to discuss the future of Linux
packaging. Topics will include RPM, the role of other packaging
technologies (dpkg, APT, yum, etc.), and the needs of ISVs in
packaging solutions".

Has this ever been discussed on Debian ML's?

Joey Hess was there. I can't recall any mention of it on the DWN.

Why do we [users] have to learn of these things from external
sources?

Regards

-- 
Ottavio Caruso

I will not purchase any computing equipment from manufacturers that recommend 
Windows Vista™ or any other Microsoft® products.
http://www.pledgebank.com/boycottvista

__
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Announce: zerotools (tools to keep virtual machine disks cleaner)

[Eduard Bloch]

#include 
* Mike Hommey [Wed, Jan 03 2007, 01:38:23PM]:
> On Wed, Jan 03, 2007 at 01:13:27PM +0100, Eduard Bloch <[EMAIL PROTECTED]> 
> wrote:
> > Please add a technical overview to your introduction. As a techie, I
> > would like to read what is so special about zerotools what I cannot do
> > with:
> > rm crapfiles && cat /dev/zero > bigfile && sync && rm bigfile && sync.
> 
> or just shred crapfiles

Not really. You need to add funny parameters like "-vzn0" to shred to
shred them with zero and then you still need to unlink them manually.
And IMO it is less intuitive than my command chain for this particular
purpose.

Eduard.

-- 
OpenBSD fails miserably in this respect, and makes for an example of how NOT
to work with the community on security issues.  Their approach is, roughly,
"we fixed this a while ago but didn't tell anyone, so you're vulnerable and
we're not, ha-ha-ha".


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Announce: zerotools (tools to keep virtual machine disks cleaner)

[Mike Hommey]

On Wed, Jan 03, 2007 at 01:13:27PM +0100, Eduard Bloch <[EMAIL PROTECTED]> 
wrote:
> Please add a technical overview to your introduction. As a techie, I
> would like to read what is so special about zerotools what I cannot do
> with:
> rm crapfiles && cat /dev/zero > bigfile && sync && rm bigfile && sync.

or just shred crapfiles

Mike


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Announce: zerotools (tools to keep virtual machine disks cleaner)

[Eduard Bloch]

#include 
* Aleksandr Koltsoff [Wed, Jan 03 2007, 05:47:36AM]:
> Hello
> 
> It was suggested that I'd post a short announcement on these two lists
> in order to get some feedback for a small set of utilities I just
> released. The documentation covers integrating the tools into Debian and
> I'd appreciate feedback on that part. Especially if I'm doing something
> which is out of line established Debian policies (hence the debian-devel
> list) [these parts are mainly listings 7 - 9 with the associated text].
> 
> I'll copy & paste the introduction text so that you may decide whether
> the program might be of interest to you. Target users for the tools are
> people who run Linux systems in a virtualized environment.
> 
> "Introduction
> 
> Zerotools are a set of tools to aid keeping virtual disks clean (by
> filling binary zero to those regions which are no longer in "use"). This

Please add a technical overview to your introduction. As a techie, I
would like to read what is so special about zerotools what I cannot do
with:
rm crapfiles && cat /dev/zero > bigfile && sync && rm bigfile && sync.

And you introduction does not tell me about anything new in your tools
and I do not like reading a long novel to learn that simple details.

Something like:

Introduction

Zerotools are a set of tools to aid keeping virtual disks clean (by
filling binary zero to those regions which are no longer in use as
visible files). In technical terms, they contain a library wrapper used
with LD_PRELOAD to make unlink calls zero-out the data before unlinking
the files and helper executables based on it.
...

Eduard.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Proposal for Lenny: Please avoid duplicated changelogs for binary packages sharing the same source package

[Tshepang Lekhonkhobe]

On 1/2/07, Tshepang Lekhonkhobe <[EMAIL PROTECTED]> wrote:

On 1/2/07, Pierre Habouzit <[EMAIL PROTECTED]> wrote:
> On Tue, Jan 02, 2007 at 04:28:59PM +0200, Tshepang Lekhonkhobe wrote:
> > On 1/2/07, Pierre Habouzit <[EMAIL PROTECTED]> wrote:
> > Why not have an extra /usr/share/doc/$source-name directory in case a
> > binary package of the same name doesn't exist or is not found to be
> > installed. This would be a problem when dependencies change, but on
> > package removal, /usr/share/doc/$source-name would be left alone if
> > there remains binary packages built from that source package.
>
>   That's what $package-common packages are for. But just to hold the
> copyright and changelog that's an overkill because those packages would
> need some kind of garbage collection somehow and that i'm quite sure
> that it would use more mirror resources than the repeated changelogs.
>
>   Well all in one it's not worth the effort, and has many drawbacks
> IMHO. I'd say that's a thing you may consider when you already need a
> $package-common for your packaging, but it's not necessarily a good
> idea.
>
>   just to know what we are talking about, on my system:
>
> du /usr/share/doc/*/changelog.Debian.gz: 16M
> du /usr/share/doc/*/copyright: 15M
> du /usr/lib/iceweasel/firefox-bin: 15M
>
>   so well, if it's not nothing, it's still quite small, and I can't
> imagine a system where 30M is an issue nowadays (well except PDA's or
> so, but here the whole /usr/share/doc is an issue at once anyway, like
> Don already explained it - on my system /usr/share/doc is 234M big).

Pretty well-justified that this would be a useless effort. Along with
Don's mention that binary packages don't have to be the same version
it just shows that my proposal sucks.

thanks


By the way, this proposal was made because I browsed through
/usr/share/doc/$package/changelog.Debian.gz to find that they were the
same for different binary packages and expecting otherwise. I guess
it's too much trouble doing separate changelogs for binary packages.

One other important note:
If a large src package builds multiple binary packages, a small change
on one of those binary packages (say OOo) which is unrelated to the
other binary packages, will cause all those other binary packages to
be rebuilt and require upgrading, which seems like a waste.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Kernel 2.6.17 and 2.6.18-3-K7 network problem

[Hendrik Sattler]

Am Mittwoch 03 Januar 2007 09:53 schrieb Antonio Laterza:
> :( ). If I choose the kernel 2.6.18 X is OK but I am not
>
> able to reache many sites. I can navigate very few sites
> example: www.google.com and www.mozilla.org. Using ethereal
> seems that the HTTP request go out (i.e. GET / ) but
> the ACK packet does not ask the next packet number  in
> other words seems that the GET request never reched the
> final site.

TCP windows scaling comes to mind:
http://lwn.net/Articles/92727/

HS


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: db.debian.org (and related infrastructure) updates

[Wouter Verhelst]

On Tue, Jan 02, 2007 at 06:32:10PM -0800, Steve Langasek wrote:
> On Wed, Jan 03, 2007 at 12:50:27AM +0100, Wouter Verhelst wrote:
> > breaking that would break software that expects this particular field to
> > be in that particular syntax.
> 
> That's not completely true; you could have an attribute type named 'gender'
> with a different OID and different syntax/semantics, you just wouldn't be
> able to use it on an LDAP server which also needed the use of the ISO
> attribute type or of any object classes that are defined to use the ISO
> attribute.

Yes; hence the quoted bit of my above paragraph.

> But if all of our Japanese, Chinese, Greek Orthodox, Muslim, and French
> Revolutionary developers can tolerate having to enter their birthdates using
> the Gregorian calendar, I think we'll be able to make do with an opt-in
> binary gender classification too.

Ack.

-- 
 Home is where you have to wash the dishes.
  -- #debian-devel, Freenode, 2004-09-22


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Announce: zerotools (tools to keep virtual machine disks cleaner)

[Amaya]

Hi, Aleksandr

Aleksandr Koltsoff wrote:
> Zerotools are a set of tools to aid keeping virtual disks clean (by
> filling binary zero to those regions which are no longer in "use").

I maintain the package perforate. It contains a very similar utility,
zum, that is just a small c binary that seems to do the same.

This package is no longer maintained upstream, so you might be
interested in looking at the source for zum.c (8 kb) and maybe tell me
if this is superseeded by your Zerotools application, which sounds very
likely to me.

Also take a look at http://bugs.debian.org/294297 and be warned that
these kind of utilities can ruin a vfat file system (just warning you
in case you have not thought about it). 

BTW, patches or a new upstream would be very welcome :)

Happy hacking!

-- 
  ·''`. If I can't dance to it, it's not my revolution
 : :' :-- Emma Goldman
 `. `'   Proudly running Debian GNU/Linux (unstable)
   `- www.amayita.com  www.malapecora.com  www.chicasduras.com


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Kernel 2.6.17 and 2.6.18-3-K7 network problem

[Antonio Laterza]

Hi all, I posted the question below to debian-user ...
maybe nobody experienced my same issue. How can I
contribute to eventually solve? Which other information are
necessary in order to focalize the problem?


I want to ask if somebody experienced the same
problem I have with etch installation and in general with
kernel 2.6.17 and 2.6.18 on AMD K7. The computer where I
tried has 2 network cards, I connect only 1 card and it
receives the IP address.

I experienced the problem installing etch using a
businesscard, using knoppix 5.0.1 (use kernel 2.6.17) and
upgrading from sarge (kernel 2.6.8) to etch (kernel
2.6.18). 
In the last case, with the same installation if I chose
kernel 2.6.8 I have no network problem (but X does not work
:( ). If I choose the kernel 2.6.18 X is OK but I am not
able to reache many sites. I can navigate very few sites
example: www.google.com and www.mozilla.org. Using ethereal
seems that the HTTP request go out (i.e. GET / ) but
the ACK packet does not ask the next packet number  in
other words seems that the GET request never reched the
final site.

I tryied to reduce the MTU but was unsuccesful. 
Last details: the network card is the Realtek 

Regards
anlater 


anlater
http://studioinglaterza.blogspot.com/


__
Do You Yahoo!?
Poco spazio e tanto spam? Yahoo! Mail ti protegge dallo spam e ti da tanto 
spazio gratuito per i tuoi file e i messaggi 
http://mail.yahoo.it 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: db.debian.org (and related infrastructure) updates

[Amaya]

Steve Langasek wrote:
> But if all of our Japanese, Chinese, Greek Orthodox, Muslim, and
> French Revolutionary developers can tolerate having to enter their
> birthdates using the Gregorian calendar, I think we'll be able to make
> do with an opt-in binary gender classification too.

ROTFL
You are so damn right! I will shut up now! :*

-- 
  ·''`. If I can't dance to it, it's not my revolution
 : :' :-- Emma Goldman
 `. `'   Proudly running Debian GNU/Linux (unstable)
   `- www.amayita.com  www.malapecora.com  www.chicasduras.com


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]