Bug#1028970: complains to stderr about "unknown extended header keyword"

2023-01-15 Thread Janos LENART
severity 1028970 wishlist
tag 1028970 wontfix
thanks

Marco, tar doesn't document a standard for stderr. In particular, this
looks like a legit warning in either case. Please bring this up with the
software that is broken by it.

On Sun, 15 Jan 2023, 16:51 Marco d'Itri,  wrote:

> Package: tar
> Version: 1.34+dfsg-1.1
> Severity: important
>
> Emitting these warnings to stderr is bad because it breaks other
> software which calls tar and does not expect something on stderr.
> Either the warnings are useless, and then they should just be
> suppressed, or they are relevant, and then tar should be taught about
> whatever a "LIBARCHIVE.xattr.com.apple.lastuseddate" is.
>
> To reproduce:
>
> $ wget
> https://cpan.metacpan.org/authors/id/M/MR/MRSCOTTY/Crypt-X509-0.55.tar.gz
> $ LANG=C tar tvvf Crypt-X509-0.55.tar.gz > /dev/null
> tar: Ignoring unknown extended header keyword
> 'LIBARCHIVE.xattr.com.apple.lastuseddate#PS'
> tar: Ignoring unknown extended header keyword
> 'LIBARCHIVE.xattr.com.apple.lastuseddate#PS'
> $
>
> --
> ciao,
> Marco
>


Bug#970717: kubernetes: abuse of "embedded code copies" | hundreds of vendored libraries

2020-09-22 Thread Janos LENART
Hi Elana,

Thank you for looking into this.

I've believed the vendor/ directory is thoroughly vetted, and I have also
checked every single directory, sometimes even files. I do not think there
is a DFSG issue here. Dmitry does not agree with the fact that vendor/ is
boundled in the Debian packages. It is not actually against the Debian
Policy either, only his interpretation of it. There was a flamewar on this
on debian-devel. Long story short, Dmitry did not maintain and did not want
to maintain the package, I've adopted it and repackaged it boundling the
vendor/ directory, Dmitry was upset because of the technical disagreement.
Nobody with actual powers had a problem with this. Marking this as a
serious bug is his latest attempt at sabotage.

Granted, the current solution is not ideal.

Janos

On Tue, Sep 22, 2020 at 6:09 PM Elana Hashman  wrote:

> On Tue, Sep 22, 2020 at 08:56:25PM +1000, Dmitry Smirnov wrote:
> >
> > As discussed in debian-devel, Kubernetes package abuses Debian practices
> > and Golang team policies by needlessly vendoring hundreds(!) of
> libraries,
> > most of which are available in Debian.
> >
> > For a complex package like Kubernetes, _some_ strategic vendoring would
> be
> > understandable for practical reasons. But not everything.
> >
> > Maintainer circumvented packaging practices and introduced re-packaged
> > Kubernetes in a state that would have never been accepted by ftp-masters.
> >
> > Please consider removing redundant libraries from "vendor".
> > In the current state, the package is unsuitable for "stable".
>
> It's not entirely clear to me if the policy concerns are around
> licensing compliance or simply the volume of vendored dependencies.
>
>
> Wearing my Kubernetes SIG Chair/upstream hat:
>
> I believe that the license compliance of everything in vendor/ has been
> thoroughly vetted, but that information may not have been adequately
> surfaced for downstream projects to use. In this case, any violations
> are surface-level/paperwork as opposed to fundamental issues with DFSG
> compliance. I've requested that upstream better surfaces this
> information in order to be able to build Kubernetes in a
> policy-compliant way in Debian:
> https://github.com/kubernetes/kubernetes/issues/94976
>
> Thanks,
>
> - e
>


-- 
LÉNÁRT, János



Bug#952222: closed by Debian FTP Masters (reply to Janos Lenart ) (Bug#952222: fixed in google-authenticator 20191231-1)

2020-04-07 Thread Janos LENART
Hi,
My mistake! Thanks for the heads up. Done.
Janos

On Tue, Apr 7, 2020 at 9:13 AM Adrian Bunk  wrote:

> On Sat, Mar 07, 2020 at 11:39:05AM +, Debian Bug Tracking System wrote:
> >...
> > Architecture: source amd64
> >...
> >* Fixed FTBFS (Closes: #95).
> >...
>
> Thanks for fixing this bug.
> Could you make an additional source-only upload so that the fix can
> migrate to testing?
>
> Thanks
> Adrian
>


-- 
LÉNÁRT, János



Bug#666129: NMU to latest git?

2013-05-29 Thread Janos LENART
In progress, sorry for the delay
google-authenticator_20130529-1_amd64.changes uploaded successfully to
localhost
along with the files:
  google-authenticator_20130529-1.dsc
  google-authenticator_20130529.orig.tar.gz
  google-authenticator_20130529-1.debian.tar.gz
  libpam-google-authenticator_20130529-1_amd64.deb

Greetings,

Your Debian queue daemon (running on host franck.debian.org)


On 27 May 2013 09:29, Mr Allwyn Fernandes  wrote:

> Hi Lenart,
>
> Just wondering if you are planning to update google-authenticator to a
> newer
> version, as per bugs #666129 and #660188, or if you mind if I do an NMU? I
> had
> built a deb from git HEAD a few months ago, and have the relevant files
> handy
> to submit if you don't have the time to look at it.
>
> Cheers,
>
> Allwyn.
>



-- 
LÉNÁRT, János



Bug#1000980: kubernetes: Please upgrade to golang-1.17

2022-02-08 Thread Janos LENART
I'll get on this. Thanks.

On Tue, 8 Feb 2022 at 17:51, Scott Kitterman  wrote:

> On Thu, 02 Dec 2021 12:29:37 +0800 Shengjing Zhu  wrote:
> > Source: kubernetes
> > Version: 1.20.5+really1.20.2-1
> > Severity: important
> > X-Debbugs-Cc: z...@debian.org
> > Control: block 998747 by -1
> >
> > Dear Maintainer,
> >
> > As part of the effort to limit the number of Go compiler in the
> > archive, please upgrade to golang-1.17.
>
> golang-1.15 is gone now, so bumped to serious.
>
> Scott K



-- 
LÉNÁRT, János