Bug#810004: jessie-pu: package php5/5.6.16+dfsg-0+deb8u1

[Salvatore Bonaccorso]

Hi Adam, hi Ondřej

On Tue, Jan 12, 2016 at 09:45:44PM +, Adam D. Barratt wrote:
> On Tue, 2016-01-12 at 20:46 +0100, Ondřej Surý wrote:
> > Hi Adam and release team,
> > 
> > sorry to bother you again with PHP :), but the inevitable has happened
> > and PHP 5.6.17+dfsg has been released with 4 security vulnerabilities.
> > Salvatore has asked me again if we could push this through SRU, since
> > you already approved 5.6.16+dfsg.
> 
> With s/jessie-security/jessie/g, please go ahead. Preferably without
> 5.6.18 appearing in the meantime :-p

Thanks for handling this via jessie-pu this time!

Regards,
Salvatore

Bug#707219: release.debian.org: check co-installability of standard

[Julien Cristau]

On Tue, Jan 12, 2016 at 10:38:05 +0100, Raphael Hertzog wrote:

> Hi,
> 
> On Wed, 08 May 2013, Julien Cristau wrote:
> > britney doesn't currently check that required+important+standard
> > packages are all co-installable.  It might be a good idea to add a
> > faux-package to check that?
> 
> This check would probably have avoided #810785 where ifupdown 0.8.6
> entered testing with a Breaks against the version of systemd which
> is in testing.
> 
> Thus it would be nice to implement this, even with a hardcoded list
> for a start, it's better than nothing.
> 
> Having debootstrap broken is a big issue for anyone working with testing.
> 
Not for "anyone", no.  Please don't pretend this is a bigger problem than
it really is.  Annoying?  Sure.  But not the end of the world, there are
easy workaround for anyone who can't deal with a few days' breakage.

Cheers,
Julien

Bug#810785: ifupdown breaks debootstrap of Debian

[Raphaël Hertzog]

Package: ifupdown
Version: 0.8.5
Severity: critical
Justification: breaks the whole system
User: de...@kali.org

debootstrap of stretch currently fails due to ifupdown:
$ sudo debootstrap --components=main,contrib,non-free stretch stretch 
http://ftp.debian.org/debian
[...]
I: Unpacking libxapian22v5...
W: Failure while installing base packages.  This will be re-attempted up
to five times.
W: See /home/rhertzog/tmp/stretch/debootstrap/debootstrap.log for details
(possibly the package /var/cache/apt/archives/ifupdown_0.8.6_amd64.deb is
at fault)

debootstrap.log contains this:

dpkg: regarding .../ifupdown_0.8.6_amd64.deb containing ifupdown:
 ifupdown breaks systemd (<< 228-3~)
  systemd (version 228-2+b1) is present and triggered.

dpkg: error processing archive /var/cache/apt/archives/ifupdown_0.8.6_amd64.deb 
(--unpack):
 installing ifupdown would break systemd, and
 deconfiguration is not permitted (--auto-deconfigure might help)



The problem is that stretch still has systemd 228-2 and 228-4 is unlikely
to migrate quickly since it just got updated and seems to have a new RC bug.

I'm not sure what's the best way forward... possibly upload something to
testing-proposed-updates to drop that breaks temporarily until the package
migrates (and ensure 0.8.7 does not migrate before systemd).

Or force migrate the new systemd despite the age and the bug...

Ccing release team and systemd maintainers to have their opinion.

-- System Information:
Debian Release: stretch/sid
  APT prefers squeeze-lts
  APT policy: (500, 'squeeze-lts'), (500, 'oldoldstable'), (500, 'unstable'), 
(500, 'testing'), (500, 'stable'), (500, 'oldstable'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.3.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=fr_FR.utf8, LC_CTYPE=fr_FR.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages ifupdown depends on:
ii  adduser  3.113+nmu3
ii  init-system-helpers  1.24
ii  iproute2 4.3.0-1
ii  libc62.21-6
ii  lsb-base 9.20150917

Versions of packages ifupdown recommends:
ii  isc-dhcp-client [dhcp-client]  4.3.3-5

Versions of packages ifupdown suggests:
ii  ppp 2.4.7-1+1
pn  rdnssd  

-- no debconf information

Bug#707219: release.debian.org: check co-installability of standard

[Raphael Hertzog]

Hi,

On Wed, 08 May 2013, Julien Cristau wrote:
> britney doesn't currently check that required+important+standard
> packages are all co-installable.  It might be a good idea to add a
> faux-package to check that?

This check would probably have avoided #810785 where ifupdown 0.8.6
entered testing with a Breaks against the version of systemd which
is in testing.

Thus it would be nice to implement this, even with a hardcoded list
for a start, it's better than nothing.

Having debootstrap broken is a big issue for anyone working with testing.

Cheers,
-- 
Raphaël Hertzog ◈ Debian Developer

Support Debian LTS: http://www.freexian.com/services/debian-lts.html
Learn to master Debian: http://debian-handbook.info/get/

Re: Bug#810785: ifupdown breaks debootstrap of Debian

[Guus Sliepen]

severity 810785 serious
thanks

On Tue, Jan 12, 2016 at 10:23:49AM +0100, Raphaël Hertzog wrote:

> Severity: critical
> Justification: breaks the whole system

While it is very annoying, it doesn't *break* the whole system. Also, it
is only a problem when debootstrapping testing, so it still works fine
in a lot of other situations.

But the problem is in any case not in the ifupdown package itself;
ifupdown has correct Breaks: headers. The package was moved to testing
because it actually is installable (although on a normal testing system,
trying to upgrade ifupdown to the latest version will cause sysvinit to
be installed and systemd removed). So I rather believe the problem is
that debootstrap doesn't handle the situation correctly:

> debootstrap.log contains this:
> 
> dpkg: regarding .../ifupdown_0.8.6_amd64.deb containing ifupdown:
>  ifupdown breaks systemd (<< 228-3~)
>   systemd (version 228-2+b1) is present and triggered.
> 
> dpkg: error processing archive 
> /var/cache/apt/archives/ifupdown_0.8.6_amd64.deb (--unpack):
>  installing ifupdown would break systemd, and
>  deconfiguration is not permitted (--auto-deconfigure might help)

Maybe debootstrap should ensure dpkg is called with --auto-deconfigure?

> The problem is that stretch still has systemd 228-2 and 228-4 is unlikely
> to migrate quickly since it just got updated and seems to have a new RC bug.
> 
> I'm not sure what's the best way forward... possibly upload something to
> testing-proposed-updates to drop that breaks temporarily until the package
> migrates (and ensure 0.8.7 does not migrate before systemd).
> 
> Or force migrate the new systemd despite the age and the bug...

I would rather see two things happening: 1) that debootstrap handles
this situation better, and 2) that packages that have Breaks: foo (<<
x.y.z) are delayed migrating to testing until foo version x.y.z or later
is also ready to migrate to testing.

But I'll see if it's possible to remove the Breaks: without really
breaking the system.

> Ccing release team and systemd maintainers to have their opinion.

-- 
Met vriendelijke groet / with kind regards,
  Guus Sliepen 


signature.asc
Description: Digital signature

Re: Bug#810785: ifupdown breaks debootstrap of Debian

[Julian Andres Klode]

On Tue, Jan 12, 2016 at 12:16:29PM +0100, Guus Sliepen wrote:
> On Tue, Jan 12, 2016 at 11:24:00AM +0100, Julian Andres Klode wrote:
> 
> > > > dpkg: error processing archive 
> > > > /var/cache/apt/archives/ifupdown_0.8.6_amd64.deb (--unpack):
> > > >  installing ifupdown would break systemd, and
> > > >  deconfiguration is not permitted (--auto-deconfigure might help)
> > > 
> > > Maybe debootstrap should ensure dpkg is called with --auto-deconfigure?
> > 
> > So, you want an unconfigured init system or other unconfigured packages in
> > the base system? Really?
> > 
> > That does not make any sense at all.
> 
> No; what happens is that apt knows that it can upgrade two packages at
> the same time, where one of them Breaks: the old version of the other
> package. But dpkg processes packages individually, and when the old
> version of the other package is still installed, and dpkg tries to
> upgrade the first package, it notices the Breaks. It can temporarily
> deconfigure the other package, continue upgrading the first, then
> upgrading the other package and (re)configuring it.

Yes it can. When *upgrading*. We are *NOT* upgrading, but bootstrapping
a new system here.

> 
> This is actually what happens when you are just doing an apt-get
> dist-upgrade on a testing system. The problem with debootstrap is that
> this --auto-deconfigure option is somehow not passed to dpkg.

We are talking about bootstrapping, not upgrading. Automatic
deconfiguration of packages is entirely irrelevant here.

Please think first.

-- 
Julian Andres Klode  - Debian Developer, Ubuntu Member

See http://wiki.debian.org/JulianAndresKlode and http://jak-linux.org/.

When replying, only quote what is necessary, and write each reply
directly below the part(s) it pertains to (`inline'). Thank you.

Bug#810568: transition: openexr

[Andreas Metzler]

On 2016-01-11 "Matteo F. Vescovi"  wrote:
> On 2016-01-11 at 20:50 (CET), Emilio Pozuelo Monfort wrote:
>> Do rdeps build fine against the new versions of the libraries?

> Well, not so fine :-/

> Following, you'll find the r-deps for OpenEXR (and IlmBase as direct
> b-dep) and IlmBase (without OpenEXR interference).
[...]
> hugin => FTBFS
[...]

Hello,

I have just tried this locally. I get a build-error with 
libopenexr-dev/experimental but get a successful build if I
additionally upgrade libilmbase-dev to experimental.

Shouldn't libopenexr-dev's dependency on libilmbase-dev be versioned,
requiring libilmbase-dev (>= 2.2.0-3)?

The dependency was versioned in 1.x.

cu Andreas
-- 
`What a good friend you are to him, Dr. Maturin. His other friends are
so grateful to you.'
`I sew his ears on from time to time, sure'

Bug#810838: RM: php-math-biginteger/1.0.2-2

[Andreas Beckmann]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: rm

Please remove php-math-biginteger from stretch. It was already removed
from unstable (#785556).
The versioned Provides in php-seclib is sufficient to install
php-horde-mapi in sid and stretch, but britney doesn't recognize this.


Andreas

Bug#810004: jessie-pu: package php5/5.6.16+dfsg-0+deb8u1

[Adam D. Barratt]

On Tue, 2016-01-12 at 20:46 +0100, Ondřej Surý wrote:
> Hi Adam and release team,
> 
> sorry to bother you again with PHP :), but the inevitable has happened
> and PHP 5.6.17+dfsg has been released with 4 security vulnerabilities.
> Salvatore has asked me again if we could push this through SRU, since
> you already approved 5.6.16+dfsg.

With s/jessie-security/jessie/g, please go ahead. Preferably without
5.6.18 appearing in the meantime :-p

Regards,

Adam

Re: Bug#810785: ifupdown breaks debootstrap of Debian

[Julian Andres Klode]

On Tue, Jan 12, 2016 at 11:01:11AM +0100, Guus Sliepen wrote:
> severity 810785 serious
> thanks
> 
> On Tue, Jan 12, 2016 at 10:23:49AM +0100, Raphaël Hertzog wrote:
> 
> > Severity: critical
> > Justification: breaks the whole system
> 
> While it is very annoying, it doesn't *break* the whole system. Also, it
> is only a problem when debootstrapping testing, so it still works fine
> in a lot of other situations.
> 
> But the problem is in any case not in the ifupdown package itself;
> ifupdown has correct Breaks: headers. The package was moved to testing
> because it actually is installable (although on a normal testing system,
> trying to upgrade ifupdown to the latest version will cause sysvinit to
> be installed and systemd removed). So I rather believe the problem is
> that debootstrap doesn't handle the situation correctly:
> 
> > debootstrap.log contains this:
> > 
> > dpkg: regarding .../ifupdown_0.8.6_amd64.deb containing ifupdown:
> >  ifupdown breaks systemd (<< 228-3~)
> >   systemd (version 228-2+b1) is present and triggered.
> > 
> > dpkg: error processing archive 
> > /var/cache/apt/archives/ifupdown_0.8.6_amd64.deb (--unpack):
> >  installing ifupdown would break systemd, and
> >  deconfiguration is not permitted (--auto-deconfigure might help)
> 
> Maybe debootstrap should ensure dpkg is called with --auto-deconfigure?

So, you want an unconfigured init system or other unconfigured packages in
the base system? Really?

That does not make any sense at all.

-- 
Julian Andres Klode  - Debian Developer, Ubuntu Member

See http://wiki.debian.org/JulianAndresKlode and http://jak-linux.org/.

When replying, only quote what is necessary, and write each reply
directly below the part(s) it pertains to (`inline'). Thank you.

Re: Bug#805549: Patch for strech, jessie, wheezy

[Felipe Sateler]

On 12 Jan 2016 03:08, "Adam D. Barratt"  wrote:
>
> On Tue, 2015-12-15 at 13:18 +, Adam D. Barratt wrote:
> > On 2015-12-15 13:04, Felipe Sateler wrote:
> > > Hi,
> > >
> > > Dear release team, I'm copying you to ask if the attached patch would
> > > be OK to upload to stable/oldstable, fixes bug #805549. This error
> > > makes some builds against stk impossible, since there are two headers
> > > missing.
> > >
> > > The version in unstable is a different upstream version, so the patch
> > > is not exactly the same (attached is a patch on a patch, currently we
> > > patch the real file, as the previous patch was accepted upstream).
> >
> > In isolation, the patch looks okay. In order to get approval for actual
> > uploads for stable and oldstable, however, please open a p-u bug for
> > each, including a full source debdiff for a package built and tested on
> > that distribution.
>
> I note that there are now #810760 and #810761. For future reference,
> "[i]n order to get approval", implied opening the bugs _before_ the
> uploads, not afterwards.

Oops. Sorry, will do correctly next time. I suppose I somehow let "built
and tested" also include "uploaded".

Saludos

Re: Bug#810785: ifupdown breaks debootstrap of Debian

[Guus Sliepen]

On Tue, Jan 12, 2016 at 11:24:00AM +0100, Julian Andres Klode wrote:

> > > dpkg: error processing archive 
> > > /var/cache/apt/archives/ifupdown_0.8.6_amd64.deb (--unpack):
> > >  installing ifupdown would break systemd, and
> > >  deconfiguration is not permitted (--auto-deconfigure might help)
> > 
> > Maybe debootstrap should ensure dpkg is called with --auto-deconfigure?
> 
> So, you want an unconfigured init system or other unconfigured packages in
> the base system? Really?
> 
> That does not make any sense at all.

No; what happens is that apt knows that it can upgrade two packages at
the same time, where one of them Breaks: the old version of the other
package. But dpkg processes packages individually, and when the old
version of the other package is still installed, and dpkg tries to
upgrade the first package, it notices the Breaks. It can temporarily
deconfigure the other package, continue upgrading the first, then
upgrading the other package and (re)configuring it.

This is actually what happens when you are just doing an apt-get
dist-upgrade on a testing system. The problem with debootstrap is that
this --auto-deconfigure option is somehow not passed to dpkg.

-- 
Met vriendelijke groet / with kind regards,
  Guus Sliepen 


signature.asc
Description: Digital signature

Re: Bug#810785: ifupdown breaks debootstrap of Debian

[Raphael Hertzog]

Hi,

On Tue, 12 Jan 2016, Michael Biebl wrote:
> I don't think there is any new open RC bug which would prevent the
> testing migration. Or am I missing your point?

No, I was just misled by the testing migration data on the tracker
that still showed one open... but it went away now as the bug has been
closed.

> So we can just wait for another 4 days and this will solve itself or we
> age the package, if it's pressing to get systemd into testing quicker.
> The latter should be fine.

Yes, aging it would be nice but at least jcristau thought that it was not
necessary and that a broken debootstrap was not worth overriding the
delay.

So I think there's not much left to discuss here.

Cheers,
-- 
Raphaël Hertzog ◈ Debian Developer

Support Debian LTS: http://www.freexian.com/services/debian-lts.html
Learn to master Debian: http://debian-handbook.info/get/

Re: Bug#810785: ifupdown breaks debootstrap of Debian

[Guus Sliepen]

On Tue, Jan 12, 2016 at 03:13:23PM +0100, Julian Andres Klode wrote:

> We are talking about bootstrapping, not upgrading. Automatic
> deconfiguration of packages is entirely irrelevant here.
> 
> Please think first.

Ok, fine, but then debootstrap shouldn't even have TRIED to install both
ifupdown and systemd, because ifupdown >= 0.8.4 and systemd << 228-3~
are EXPLICITLY not coinstallable.

-- 
Met vriendelijke groet / with kind regards,
  Guus Sliepen 


signature.asc
Description: Digital signature

Re: Bug#810785: ifupdown breaks debootstrap of Debian

[Julian Andres Klode]

On Tue, Jan 12, 2016 at 03:38:33PM +0100, Guus Sliepen wrote:
> On Tue, Jan 12, 2016 at 03:13:23PM +0100, Julian Andres Klode wrote:
> 
> > We are talking about bootstrapping, not upgrading. Automatic
> > deconfiguration of packages is entirely irrelevant here.
> > 
> > Please think first.
> 
> Ok, fine, but then debootstrap shouldn't even have TRIED to install both
> ifupdown and systemd, because ifupdown >= 0.8.4 and systemd << 228-3~
> are EXPLICITLY not coinstallable.

Yes, that is exactly the point.


-- 
Julian Andres Klode  - Debian Developer, Ubuntu Member

See http://wiki.debian.org/JulianAndresKlode and http://jak-linux.org/.

When replying, only quote what is necessary, and write each reply
directly below the part(s) it pertains to (`inline'). Thank you.

** Misc-Home/Radon Inspector

[Brochure Designs Included!]

Home/Radon Professional Inspector

We'll Custom Design Your Company Brochures, Postcards, 
Business Cards, etc., At NO CHARGE!

ALSO, SAVE 40-60% ON FULL-COLOR PRINTING!

Go To  www.PerceptiveGraphics.com  to 
request a SAMPLE KIT 
and see some Home Inspector examples!

Example: 8.5x11 Brochures: $199.95/500, $249.95/1,000, $349.95/2,500, 
$399.95/5,000

More Sizes & Styles Available! 

Proudly Made In The USA! 

Please Advise, Thank You, 
John Pitoniak, Sr. Partner, National Product Development  
References Available

P.S. If you find this email in error, we apologize. Please reply with "Remove" 
in the subject line and we will take care that you do not receive any further 
mail. Thank you.

%Date()%