etch release policy: scripts must set appropriate #! line? (was: lintian: false executable-not-elf-or-script warning with perl magic header)

2006-04-19 Thread Frank Küster
Hi release team, this discussion is about scripts with a magic header instead of a standard shebang line. It turned out that some magic headers, e.g. the one suggested in perldoc perlrun, only work if there is *no* shebang line. I think the Etch RC policy should be clarified or changed, see

Re: Getting rid of ucd-snmp (libsnmp4.2)

2006-04-19 Thread Pierre Habouzit
Le Mer 19 Avril 2006 16:37, Jochen Friedrich a écrit : Hi Pierre, uploaded in delayed/2 with urgency set to high. I just applied the patch mentioned twice on the BTS. Currently, cheops is the only app still using this ancient SNMP library (at least on i386, cheops uses libsnmp9 on all

has RC bugs bogusly

2006-04-19 Thread Thomas Bushnell BSG
According to the excuses for testing, one reason gnucash can't go in is because it has release critical bugs. But the only RC bug against gnucash is for the version in *experimental* and does not apply to the version in unstable. It is unreasonable, I think, for bugs reported only against the

Re: has RC bugs bogusly

2006-04-19 Thread Julien Danjou
On Wed, Apr 19, 2006 at 08:58:53AM -0700, Thomas Bushnell BSG wrote: According to the excuses for testing, one reason gnucash can't go in is because it has release critical bugs. But the only RC bug against gnucash is for the version in *experimental* and does not apply to the version in

Re: 'has RC bugs' bogusly

2006-04-19 Thread Thijs Kinkhorst
On Wed, April 19, 2006 18:09, Julien Danjou wrote: It is unreasonable, I think, for bugs reported only against the experimental version to keep packages out of testing. This bug should have been tagged 'experimental', I guess, and so will would have been ignored. I tagged it. You should use

Re: 'has RC bugs' bogusly

2006-04-19 Thread Pierre Habouzit
Le Mer 19 Avril 2006 18:13, Thijs Kinkhorst a écrit : On Wed, April 19, 2006 18:09, Julien Danjou wrote: It is unreasonable, I think, for bugs reported only against the experimental version to keep packages out of testing. This bug should have been tagged 'experimental', I guess, and so

Re: 'has RC bugs' bogusly

2006-04-19 Thread Andreas Barth
* Thijs Kinkhorst ([EMAIL PROTECTED]) [060419 18:13]: On Wed, April 19, 2006 18:09, Julien Danjou wrote: It is unreasonable, I think, for bugs reported only against the experimental version to keep packages out of testing. This bug should have been tagged 'experimental', I guess, and so

Re: has RC bugs bogusly

2006-04-19 Thread Thomas Bushnell BSG
Julien Danjou [EMAIL PROTECTED] writes: This bug should have been tagged 'experimental', I guess, and so will would have been ignored. I see; thanks for this. This method of course predates the nice versioning of bugs in the BTS. Now that we have versioning, it should be possible to handle

Re: has RC bugs bogusly

2006-04-19 Thread Thomas Bushnell BSG
Julien Cristau [EMAIL PROTECTED] writes: It is unreasonable, I think, for bugs reported only against the experimental version to keep packages out of testing. Other than this one bug, I can't see why gnucash is out of testing, as well, though perhaps I'm misunderstanding something about the

Re: dropbear (0.47-1 to 0.48-1) testing migration

2006-04-19 Thread Gerrit Pape
On Thu, Apr 13, 2006 at 10:39:25PM -0700, Steve Langasek wrote: On Thu, Apr 13, 2006 at 09:36:40AM +, Gerrit Pape wrote: Hi, why can't the dropbear package 0.48-1 migrate to testing? Bug #357323 applies to version 0.45-2sarge0 only, if at all, it seems. Because there's

Re: has RC bugs bogusly

2006-04-19 Thread Andreas Barth
* Thomas Bushnell BSG ([EMAIL PROTECTED]) [060419 18:29]: This method of course predates the nice versioning of bugs in the BTS. Now that we have versioning, it should be possible to handle this automatically, right? Yes. See http://lists.debian.org/debian-release/2005/08/msg00050.html for a

Getting rid of ucd-snmp (libsnmp4.2)

2006-04-19 Thread Jochen Friedrich
Hi Pierre, uploaded in delayed/2 with urgency set to high. I just applied the patch mentioned twice on the BTS. Currently, cheops is the only app still using this ancient SNMP library (at least on i386, cheops uses libsnmp9 on all other archs and in Ubuntu, as well). I'd really prefer to

Patch for update-grub for Sarge R3

2006-04-19 Thread Manoj Srivastava
Hi, The /sbin/update-grub script is often used in the postinst hook of kernel images, indeed, the man page for update-grub(8) explains how to hook the script in. Even in Sarge, all user interaction is meant to be done via debconf; and debconf does not like maintainer scripts sendng