Re: cgiirc Hijacking

2006-06-21 Thread Mario 'BitKoenig' Holbe
Joe Smith [EMAIL PROTECTED] wrote:
 As I understand it, there is no good reason to have s.d.o in
 my sources list, as the packages in there are for sarge, and may not be
 compatible with the current sid ABI.

This is nonsense. If this should really be the way you understand it,
please ask yourself why a package's version on s.d.o which overrides a
version in unstable (i.e. the version on s.d.o is bigger than the
version in unstable) should ever have a less compatible ABI than the
(smaller) version in unstable.


regards
   Mario
-- 
It is a capital mistake to theorize before one has data.
Insensibly one begins to twist facts to suit theories instead of theories
to suit facts.   -- Sherlock Holmes by Arthur Conan Doyle


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



Re: cgiirc Hijacking

2006-06-21 Thread Jeroen van Wolffelaar
On Wed, Jun 21, 2006 at 08:07:28AM +0200, Mario 'BitKoenig' Holbe wrote:
 Joe Smith [EMAIL PROTECTED] wrote:
  As I understand it, there is no good reason to have s.d.o in
  my sources list, as the packages in there are for sarge, and may not be
  compatible with the current sid ABI.
 
 This is nonsense. If this should really be the way you understand it,
 please ask yourself why a package's version on s.d.o which overrides a
 version in unstable (i.e. the version on s.d.o is bigger than the
 version in unstable) should ever have a less compatible ABI than the
 (smaller) version in unstable.

You should not mix suites (releases) in your sources.list generally,
espcially not stable with testing/unstable. Security.d.o for stable
might have packages that are no longer present in testing/unstable,
which would make it undesirable to install the security.d.o versions,
also, if there's something really worthwhile in security.d.o for stable,
that should also be made available in appropriate form for
testing/unstable. It's the job of the maintainer(s) to oversee this, and
ensure that it happens.

There is no reason a user should (need to) add stable security for
his/her unstable machine.

Elsewhere in this thread there's already discussion about the technical
details why it didn't happen yet in this case and how it should happen,
I'm not repeating that discussion here.

--Jeroen

-- 
Jeroen van Wolffelaar
[EMAIL PROTECTED] (also for Jabber  MSN; ICQ: 33944357)
http://Jeroen.A-Eskwadraat.nl


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



Re: cgiirc Hijacking

2006-06-21 Thread Florian Weimer
* Mario Holbe:

 We did. 0.5.4-6sarge1 was on s.d.o as soon as possible. Since there were
 no newer version in unstable, the version on s.d.o should have had
 automatically override even the unstable version. Of course, if you
 don't source in s.d.o, you don't get security updates :)

In this case, the security update should have been propagated to
testing and unstable automatically.  For a few months, dak behaved
that way.  I don't know what has changed.


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



Re: cgiirc Hijacking

2006-06-21 Thread Martin Zobel-Helas
Hi Florian,

On Wed, Jun 21, 2006 at 07:34:49PM +0200, Florian Weimer [EMAIL PROTECTED] 
wrote:

 * Mario Holbe:
 
  We did. 0.5.4-6sarge1 was on s.d.o as soon as possible. Since there were
  no newer version in unstable, the version on s.d.o should have had
  automatically override even the unstable version. Of course, if you
  don't source in s.d.o, you don't get security updates :)
 
 In this case, the security update should have been propagated to
 testing and unstable automatically.  For a few months, dak behaved
 that way.  I don't know what has changed.

For proposed-updates is a new queue-handling in place [1]. This means
that all packages which get uploaded to security.debian.org get into
this new proposed-update queue, before they get accepted into the real
proposed-update queue. For dak packages in the new proposed-update queue
are just as _not being uploaded at all_ before they are accepted by the
stable release team members.

cgiirc was a special case as version in sid and version in sarge were
the same. Packages from security.d.o still get synced to ftpmaster.d.o,
but into the new proposed-queue, which needs interaction by the stable
release managers. 

We are aware of this, and it should be solved (hopefully) by the next
dinstall run.

Greetings
Martin

[1] http://lists.debian.org/debian-devel-announce/2006/06/msg7.html


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



Re: cgiirc Hijacking

2006-06-20 Thread Elrond
On Mon, Jun 19, 2006 at 07:02:50PM -0300, Damián Viano wrote:
 Hi, 
 
   I've seen cgiirc[1] in a bad state for some time now, I tried to contact
 the maintainer (Mario Holbe) more than a month ago, offering my help and
 my work[2]. No answer so far.

Mario and I were busy. And I was trying to update the packaging.

Our sponsor is on vacation, we planned to upload when he
will return.


   This is a sponsored uploaded package that only had 1 upload, and had a
 DSA[3] issued which is still not fixed outside of sarge.
[...]

We made that DSA happen.

And we were told more than once, that the sarge fix would
propagate to unstable und ultimately testing. This did not
happen. Not our fault.


   It has been removed from testing and it would be a shame to ship
 without it, since it's just a maintenance problem.

packages.debian.org/cgiirc has it in testing.


[...]
   I currently have a developer willing to sponsor my work on this
 package, we'll be uploading tonight after the daily dak run if there is
 no further news on/from Mario. The urgency is mostly for the security
 bug.

Mario and I are happy, that you're now maintaining this
package. It frees us for other open source work we have to
care for and our real life.


As I still have some stuff for cgiirc packaging in my local
trees (partly by Paul Wise), I will file bugs with patches,
so you can consider including them.


Elrond


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



Re: cgiirc Hijacking

2006-06-20 Thread Margarita Manterola

On 6/20/06, Elrond [EMAIL PROTECTED] wrote:

On Mon, Jun 19, 2006 at 07:02:50PM -0300, Damián Viano wrote:
   I've seen cgiirc[1] in a bad state for some time now, I tried to contact
 the maintainer (Mario Holbe) more than a month ago, offering my help and
 my work[2]. No answer so far.
Mario and I were busy. And I was trying to update the packaging.
Our sponsor is on vacation, we planned to upload when he will return.


In cases where a security bug is being fixed, you usually try to
upload the package as soon as possible.  If your sponsor is on
vacation, you should find another sponsor.  If you have trouble
preparing the package, then ask for help... But not let the bug sit
unfixed for more than a month.


   This is a sponsored uploaded package that only had 1 upload, and had a
 DSA[3] issued which is still not fixed outside of sarge.
We made that DSA happen.
And we were told more than once, that the sarge fix would
propagate to unstable und ultimately testing. This did not
happen. Not our fault.


Who told you that the sarge fix would propagate?

Packages don't *propagate* from stable.  If you want a package that
was uploaded to stable to go to unstable, an upload is needed.  You
should have asked for a sponsor.



Elrond


This is my personal opinion, but I know that other people share it: it
would be really nice if you could use your real name in what regards
to Debian communication.  Nicknames are accepted on IRC, but when
dealing with packages, bugs, and similiar stuff, we all stick to our
real names.

--
Besos,
Marga



Re: cgiirc Hijacking

2006-06-20 Thread Mario 'BitKoenig' Holbe
On Tue, Jun 20, 2006 at 01:18:11PM -0300, Margarita Manterola wrote:
 In cases where a security bug is being fixed, you usually try to
 upload the package as soon as possible.  If your sponsor is on

We did. 0.5.4-6sarge1 was on s.d.o as soon as possible. Since there were
no newer version in unstable, the version on s.d.o should have had
automatically override even the unstable version. Of course, if you
don't source in s.d.o, you don't get security updates :)

 preparing the package, then ask for help... But not let the bug sit
 unfixed for more than a month.

We didnt.


Mario
-- 
There is nothing more deceptive than an obvious fact.
 -- Sherlock Holmes by Arthur Conan Doyle


signature.asc
Description: Digital signature


Re: cgiirc Hijacking

2006-06-20 Thread Joe Smith


Mario 'BitKoenig' Holbe [EMAIL PROTECTED] wrote in message 
news:[EMAIL PROTECTED]



On Tue, Jun 20, 2006 at 01:18:11PM -0300, Margarita Manterola wrote:

In cases where a security bug is being fixed, you usually try to
upload the package as soon as possible.  If your sponsor is on



We did. 0.5.4-6sarge1 was on s.d.o as soon as possible. Since there were
no newer version in unstable, the version on s.d.o should have had
automatically override even the unstable version. Of course, if you
don't source in s.d.o, you don't get security updates :)


I run unstable and do not have s.d.o
As I understand it, there is no good reason to have s.d.o in
my sources list, as the packages in there are for sarge, and may not be
compatible with the current sid ABI.

Besides, s.d.o is already a highly stressed server. (AFAIK) 




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



Re: cgiirc Hijacking

2006-06-20 Thread Thijs Kinkhorst
On Tue, 2006-06-20 at 13:18 -0300, Margarita Manterola wrote:
 Who told you that the sarge fix would propagate?
 
 Packages don't *propagate* from stable.  If you want a package that
 was uploaded to stable to go to unstable, an upload is needed.  You
 should have asked for a sponsor.

Well, at least this used to work in the past. If the version in stable
was greater than that in unstable or testing, that version would also
propagate there. This is not only convenient for security updates to
packages with the same version in stable as in unstable, but also makes
sure the condition stable = testing = unstable remains valid.

Appearently this didn't happen here, but as far as I understand it,
that's a bug.


Thijs


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


Re: cgiirc Hijacking

2006-06-20 Thread Steve Langasek
On Tue, Jun 20, 2006 at 01:18:11PM -0300, Margarita Manterola wrote:
This is a sponsored uploaded package that only had 1 upload, and 
 had a
  DSA[3] issued which is still not fixed outside of sarge.
 We made that DSA happen.
 And we were told more than once, that the sarge fix would
 propagate to unstable und ultimately testing. This did not
 happen. Not our fault.

 Who told you that the sarge fix would propagate?

 Packages don't *propagate* from stable.

Yes, they most certainly do... when the proposed-updates queue is fully
operational, which it has not been since the last point release of sarge.

-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
[EMAIL PROTECTED]   http://www.debian.org/


signature.asc
Description: Digital signature


Re: cgiirc Hijacking

2006-06-20 Thread Jeroen van Wolffelaar
On Tue, Jun 20, 2006 at 10:45:27PM +0200, Thijs Kinkhorst wrote:
 On Tue, 2006-06-20 at 13:18 -0300, Margarita Manterola wrote:
  Who told you that the sarge fix would propagate?
  
  Packages don't *propagate* from stable.  If you want a package that
  was uploaded to stable to go to unstable, an upload is needed.  You
  should have asked for a sponsor.
 
 Well, at least this used to work in the past. If the version in stable
 was greater than that in unstable or testing, that version would also
 propagate there. This is not only convenient for security updates to
 packages with the same version in stable as in unstable, but also makes
 sure the condition stable = testing = unstable remains valid.
 
 Appearently this didn't happen here, but as far as I understand it,
 that's a bug.

The package isn't in sarge/stable on ftp-master/all mirrors, only on
security.d.o, that's why. Not a bug, but a 'feature' -- the package
hasn't been approved yet for stable[1], so neither propagated to
testing/unstable.

--Jeroen

[1] Due to infractructure not being ready yet, mostly

-- 
Jeroen van Wolffelaar
[EMAIL PROTECTED] (also for Jabber  MSN; ICQ: 33944357)
http://Jeroen.A-Eskwadraat.nl


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



cgiirc Hijacking

2006-06-19 Thread Damián Viano
Hi, 

I've seen cgiirc[1] in a bad state for some time now, I tried to contact
the maintainer (Mario Holbe) more than a month ago, offering my help and
my work[2]. No answer so far.

This is a sponsored uploaded package that only had 1 upload, and had a
DSA[3] issued which is still not fixed outside of sarge.

Several upstream versions have passed unnoticed.

It has been removed from testing and it would be a shame to ship
without it, since it's just a maintenance problem.

I currently use[4] and maintain[5] this package and think that if for
whatever reason the actual maintainer can't keep it in good shape my
work could very well be useful for many other debian users. Therefore
my intention to hijack this package.

I currently have a developer willing to sponsor my work on this
package, we'll be uploading tonight after the daily dak run if there is
no further news on/from Mario. The urgency is mostly for the security
bug.

Damián Viano(Des).

[1] http://packages.qa.debian.org/cgiirc 
[2] http://damianv.com.ar/debian/cgiirc 
[3] http://www.debian.org/security/2006/dsa-1052
[4] http://irc.lug.fi.uba.ar/
[5] http://damianv.com.ar/cgi-bin/packages.cgi


signature.asc
Description: Digital signature