Give-backs of linux/mips{,64}el / bookworm this morning

[Adam D. Barratt]

Hi Adrian,

While checking the status of the linux upload to proposed-updates this
morning, I noticed that the packages for mipsel and mips64el were still
in the BD-Uninstallable state, whereas those for some other
architectures had already started building.

Further investigation showed that you requested a give-back for the
package on those architectures while they were in Needs-Build. Could
you please explain why you did so?

For (o-)p-u, this is particularly disruptive, as packages only
automatically transition from BD-Uninstallable to Needs-Build at each
dinstall. If I hadn't spotted the situation and intervened, we would
have had at least a further 6 hour delay before either of the builds
started.

This would be annoying at the best of times, but with the freeze for
6.1 coming this weekend was even more so in this instance.

Regards,

Adam

Re: Transition unstable->testing of two packages

[Adam D. Barratt]

Hi,

On Wed, 2021-06-02 at 15:50 +0200, Filippo Rusconi wrote:
> Greetings, Fellow Developers,
> 
> I would like in this mail to advocate the unblocking of the
> transition for two
> packages:
> 

The debian-wb-team list is related to the buildd network, not testing
migration.

If you want to request that your packages be unblocked, you'd need to
file a bug against release.debian.org (preferably by using "reportbug
release.debian.org" and selecting the "unblock" option, so that it has
the appropriate content and metadata).

Regards,

Adam

Re: key chopped off

[Adam D. Barratt]

On Wed, 2020-04-08 at 20:40 +0200, Hector Oron wrote:
> - ftpmaster +wanna-build
> 
> Missatge de Jim Sughrue  del dia dc., 8
> d’abr.
> 2020 a les 16:43:
> > Hi,
> > 
> > The key for graphs here is chopped off such that the graph can't be
> > interpreted: https://buildd.debian.org/stats/
> > 
> > Could this be fixed?
> > 
> >  Jim Sughrue
> 
> Kurt or someone familiar with graphs can take a look?

That would be https://bugs.debian.org/954872

Regards,

Adam

Re: node-yarnpkg is stuck in Uploaded state for 50+ days

[Adam D. Barratt]

On Fri, 2020-02-07 at 18:50 +0100, Cyril Brulebois wrote:
> Hi,
> 
> Andreas Beckmann  (2020-02-06):
> > please do the right thing to help node-yarnpkg out of the Uploaded
> > state
> > in sid where it is for 50+ days already.
> > 
> > Should there be some process that automatically reports stale
> > Uploaded
> > packages after x days?
> 
> It seems to have reached dak just fine, but:
> 
> kibi@coccia:~$ xzgrep node-yarnpkg /srv/ftp-
> master.debian.org/log/2019-12.xz 
> 20191217121905|process-upload|dak|Processing changes file|node-
> yarnpkg_1.21.1-1_source.changes
> 20191217121933|process-upload|dak|REJECT|node-yarnpkg_1.21.1-
> 1_source.changes
> 20191217144608|process-upload|dak|Processing changes file|node-
> yarnpkg_1.21.1-1_source.changes
> 20191217144659|process-upload|dak|ACCEPT|node-yarnpkg_1.21.1-
> 1_source.changes
> 20191217144704|process-upload|dak|Archiving|node-yarnpkg_1.21.1-
> 1_source.buildinfo
> 20191217151914|process-upload|dak|Processing changes file|node-
> yarnpkg_1.21.1-1_all.changes
> 20191217151914|process-upload|dak|REJECT|node-yarnpkg_1.21.1-
> 1_all.changes
> 20191219145045|manage-build-queues|dak|removed source from build
> queue|buildd-unstable|node-yarnpkg|1.21.1-1

The initial source REJECT was due to the upload apparently trying to
add new packages while being uploaded by a DM. The subsequent arch:all
reject was:

yarnpkg_1.21.1-1_all.deb: has 2 file(s) with a timestamp too far in the
past:
  usr/share/nodejs/yarn/node_modules/npm-logical-tree/index.js (Thu
Jan  1 00:00:00 1970)  usr/share/nodejs/yarn/node_modules/npm-logical-
tree/package.json (Thu Jan  1 00:00:00 1970)

So this might need fixing on the package side.

Regards,

Adam

Re: Please remove extra Dep-Wait on mipsel

[Adam D. Barratt]

On Fri, 2019-12-27 at 11:54 +0100, Pino Toscano wrote:
> because of the update of libglvnd/mesa/qt5, some packages failed to
> build, and were given a Dep-Wait on the newer qtbase-opensource-src.
> The fixed version was already built and installed yesterday, however
> the packages are still stuck with the Dep-Wait. Can you please remove
> it for the following packages:
> - trace-cmd
> - gmic
> - kate
> - sigil
> - vtk-dicom
> - seafile-client

Done.

Regards,

Adam

Re: Crypto++ 8.0 Armel failed build

[Adam D. Barratt]

On 2019-01-30 13:19, Jeffrey Walton wrote:

Hi Everyone,

This may be an opportunity for improvement in the build logs. I'm a
fellow who helps maintain a library (Crypto++), which is packaged by a
Debian maintainer (László).


As a side note, "Debian maintainer" (usually with an upper-case "M") 
refers to a specific status, which László does not hold, as evidenced by 
his @debian.org address - he's an uploading Debian Developer.



I'm not familiar with the Debian build machines, so the additional
information would be helpful to me during troubleshooting:


I wonder how much of this information is already present or trivially 
derivable.



By the way, the script outputs:

+--+
| Build environment|
+--+

Kernel: Linux 4.9.0-8-armmp-lpae armhf (armv7l)
Toolchain package versions: binutils_2.31.1-11
...

That should probably include a 'gcc --version', ' gcc -dumpmachine'
output, 'uname -a' and 'uname -m'.


It already does include "uname -m". The "kernel" link above is "sysname 
release version build-arch (machine)", with all but "build-arch" coming 
direct from uname - see 
https://sources.debian.org/src/sbuild/0.78.0-2/lib/Sbuild/ResolverBase.pm/?hl=791#L791


The "toolchain package versions" line that was snipped also already 
includes "g++-8_8.2.0-14 gcc-8_8.2.0-14", which includes the useful 
parts of the information you're going to get from "gcc --version"


Regards,

Adam

Re: buildd status page down

[Adam D. Barratt]

On Sat, 2017-08-12 at 01:14 +0900, Roger Shimizu wrote:
> On Sat, Aug 12, 2017 at 12:09 AM, Roger Shimizu  
> wrote:
> > To whom may concern,
> >
> > I just noticed that buildd status page doesn't show properly as usual,
> > for example:
> > - https://buildd.debian.org/status/package.php?p=linux
> >
> > The error messageis:
> >   Connection to the PGdb failed!
> 
> I see actually the service should be maintained by wanna build team.
> so adding the address in loop.

This is a side-effect of the Bytemark issues mentioned in
https://lists.debian.org/debian-infrastructure-announce/2017/08/msg0.html

Regards,

Adam

Re: Reschedule a build of gdk-pixbuf?

[Adam D. Barratt]

On Sat, 2016-12-24 at 09:00 +0100, Petter Reinholdtsen wrote:
> Hi.  Can you reschedule a build of gdk-pixbuf on mips64el?  We have
> tried to reproduce the failure reported in
> https://bugs.debian.org/849136 > without success.

Looks like someone did that yesterday evening.

Regards,

Adam

Re: Please update pbuilder environments on build daemons [Was: Source upload of r-cran-treescape does not build on any architecture - but why?]

[Adam D. Barratt]

On 2016-12-21 11:42, Emilio Pozuelo Monfort wrote:

On 21/12/16 12:39, Andreas Tille wrote:

Hi,

as explained below auto builders should either have dpkg 1.18.15 or
1.18.17 (see bug #848422).


The chroots will be updated in a few hours.


It's also generally safe to assume that the relevant teams are aware of 
things like widespread dpkg and buildd breakage.


[...]
Is bach.hen...@gmail.com the correct address for "contacting 
wanna-build

people"?  If yesm Henrik is in CC - if not what's the proper contact?


I have no idea who that is, but it's not an address I've ever seen 
referenced in relation to wanna-build.


Regards,

Adam

Re: check status of tsocks on !linux

[Adam D. Barratt]

On Sat, 2016-02-27 at 19:26 +0100, Samuel Thibault wrote:
> Hello,
> 
> Andreas Beckmann, on Sat 27 Feb 2016 15:58:29 +0100, wrote:
> > could someone check the status of tsocks on hurd and kfreebsd, the
> > packages are listed as "Uploaded" for "41 days". Something seems to need
> > a little shaking to get them "Installed" :-)
> > 
> > https://buildd.debian.org/status/package.php?p=tsocks=unstable
> 
> This is what I get in the REJECT ftpmaster mail:
> 
> > tsocks-dbgsym_1.8beta5-9.3_hurd-i386.deb: APT could not parse Built-Using 
> > field
> 
> I don't know why tsocks is particular here.

According to the hurd-i386 log at
https://buildd.debian.org/status/fetch.php?pkg=tsocks=hurd-i386=1.8beta5-9.3=1456586414
 :

 Built-Using: libc6 (= )

The changelog says:

   * Document with Built-Using that tsocks ships the saveme binary,
 that is statically linked against libc6 (Closes: #769343).

so I assume the patch is failing to take account of libc having
different names across architectures. Indeed:

EXTRA_PACKAGES="libc6"
echo -n "extra:Built-Using=" >> debian/tsocks.substvars
for package in $EXTRA_PACKAGES; do
  dpkg-query -f '${source:Package} (= ${source:Version}), ' -W $package 
2>/dev/null
done >> debian/tsocks.substvars

CC-ing the NMUer, as this will need fixing in the package.

Regards,

Adam

Re: binNMUs: please exercise some care

[Adam D. Barratt]

On 2015-10-23 12:02, Thorsten Glaser wrote:

On Fri, 23 Oct 2015, Adam D. Barratt wrote:

wanna-build does, yes, but at least the Release Team tend to use the 
"wb"
wrapper tool which automatically works out the next free number on 
each

architecture.


Ah, cool – so we have only to patch this tool to automatically
use the highest number per batch on all affected architectures
(or even to use the highest number if all architectures would
be touched, but that’s probably an unreasonable amount of code
change).


Well, except you only really want to do it for libraries that are 
ma:same, as that's the only case where it actually matters and otherwise 
you're pointlessly losing versions.


It's also not quite that simple, even working things out by hand - see 
#599128 for example.



Where’s the source code to that tool?


http://anonscm.debian.org/cgit/debian-release/release-tools.git/ (in 
scripts/).


Regards,

Adam

Re: dinstall trigger problems?

[Adam D. Barratt]

[CC += ftpmaster]

On Sun, 2015-08-30 at 21:57 +0100, Adam D. Barratt wrote:
 It looks like wanna-build updates for suites other than sid and
 experimental broke on Saturday morning.
 
 Looking on wuiet, the mirror pushes seem to be working okay (as
 archive/debian/archive is up-to-date) but something is apparently
 breaking after that - for instance:
 
 -rw-r--r-- 1 wbadm wbadm 800659 Aug 29
 03:56 /srv/wanna-build/tmp/archive/debian/workdir/Packages.jessie.amd64
 
 The merge logs looked okay from a quick glance and I couldn't spot any
 other logs that looked relevant.

Looking on ftp-master after a comment made elsewhere, it looks like this
might be a mirroring issue after all, as the archvsync logs also haven't
updated since Saturday morning.

Regards,

Adam

dinstall trigger problems?

[Adam D. Barratt]

Hi,

It looks like wanna-build updates for suites other than sid and
experimental broke on Saturday morning.

Looking on wuiet, the mirror pushes seem to be working okay (as
archive/debian/archive is up-to-date) but something is apparently
breaking after that - for instance:

-rw-r--r-- 1 wbadm wbadm 800659 Aug 29
03:56 /srv/wanna-build/tmp/archive/debian/workdir/Packages.jessie.amd64

The merge logs looked okay from a quick glance and I couldn't spot any
other logs that looked relevant.

Regards,

Adam

Re: dinstall trigger problems?

[Adam D. Barratt]

On Sun, 2015-08-30 at 22:09 +0100, Adam D. Barratt wrote:
 [CC += ftpmaster]
 
 On Sun, 2015-08-30 at 21:57 +0100, Adam D. Barratt wrote:
  It looks like wanna-build updates for suites other than sid and
  experimental broke on Saturday morning.
[...]
 Looking on ftp-master after a comment made elsewhere, it looks like this
 might be a mirroring issue after all, as the archvsync logs also haven't
 updated since Saturday morning.

Joerg has tracked this down to a recent patch which was merged in dak
and hopefully fixed it; thanks.

Regards,

Adam

Re: Please give-back python-cffi on arm64

[Adam D. Barratt]

On Wed, 2015-07-08 at 22:25 +0200, Stefano Rivera wrote:
 python-cffi FTBFS on arm64 because of a libffi bug (#785756). A version
 fixing that bug just hit unstable.
 
 So, please,
 
  gb python-cffi_1.1.2-1 . arm64

Done

  gb python-cffi_1.1.2-2 . arm64
 
 (The second one is for an upload in experimental)

In that case you need to specify the distribution:

gb python-cffi_1.1.2-2 . arm64 . experimental

Done that.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1436388370.19148.6.ca...@adam-barratt.org.uk

Re: Bug#783404: jessie-pu: package perl/5.20.2-3+deb8u1

[Adam D. Barratt]

On Sun, 2015-05-03 at 16:05 +0100, Dominic Hargreaves wrote:
[...]
 This doesn't seem to have built:
 
 https://buildd.debian.org/status/package.php?p=perlsuite=jessie
 
 This appear to be related to the change in 5.20.2-2 to make perl depend on
 a matching perl-base (see #779455). However this doesn't seem to break
 builds in sid[1], which I don't quite understand; maybe a timing issue
 over when the the new perl-modules package enters the archive compared
 with the p-u suites?
 
 I'm not quite sure what the correct fix is here; maybe someone more
 familiar with this part of the infrastructure can suggest something?

While people ponder what the difference between the sid and p-u setups
is that causes this (or, more precisely, doesn't cause it in sid), as an
experiment we've told wanna-build to ignore the installability checker
issue on amd64 (as the versioned dependency from perl to perl-modules is
otherwise met in Jessie) and try building the package anyway. If that
works then we can try other architectures as well.

It's not an ideal long-term solution but will hopefully at least unblock
the logjam.

Regards

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/1430668707.7652.18.ca...@adam-barratt.org.uk

libgcrypt20 binNMUs

[Adam D. Barratt]

Hi,

I scheduled binNMUs for libgcrypt20 a few days ago, but something odd 
seems to be going on.


The binNMUs claim to have been building on all architectures for several 
days - rather than the expected few minutes - but in each case I checked 
the buildd in question had built several other packages in the meantime. 
I gave back the amd64 build as a test; afterwards Julien Cristau checked 
the log on binet and reported that log says binet took it and then 
decided it had nothing to do.


Please could someone have a look?

Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/5aa96f5ccaad579ccb9a9278331cb...@mail.adsl.funky-badger.org

Re: libgcrypt20 binNMUs

[Adam D. Barratt]

On 2014-12-02 11:21, Samuel Thibault wrote:

Adam D. Barratt, le Tue 02 Dec 2014 11:03:57 +, a écrit :
I scheduled binNMUs for libgcrypt20 a few days ago, but something odd 
seems

to be going on.

The binNMUs claim to have been building on all architectures for 
several

days -

[...]

YAML::Tiny found illegal characters in plain scalar: 'rebuild against
libgpg-error = 1.17-1 for stricter package dependency. Closes:
#771100' at /usr/share/perl5/Buildd/Daemon.pm line 354.

It seems sbuild doesn't like some characters there. After a few tries,
it seems it's the ':' character which poses problem.


Adjusting the message indeed fixes things; thanks.

Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/e028ee433a148447f37ccd0de9cda...@mail.adsl.funky-badger.org

Re: Archive mirror in /srv/wanna-build/tmp/archive at wuiet

[Adam D. Barratt]

On 2014-08-05 12:12, Joachim Breitner wrote:

I may be mistaken, but I have the impression that the archive mirror
at /srv/wanna-build/tmp/archive on wuiet is no longer updated (last
update 20:00 yesterday. I *believe* it used to be updated more than 
once

per day).


As an additional data point, there are several rsync processes running 
which all started at 01:25 and:


-r--r--r-- 1 wbadm wbadm 1 Aug  5 01:25 /srv/wanna-build/triggers/lock

Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/96948c14ab02dc34cdaf7d269d9c4...@mail.adsl.funky-badger.org

Re: P-a-s for stable and stable-security

[Adam D. Barratt]

On Sun, 2014-07-27 at 00:22 +0200, Philipp Kern wrote:
 On Sat, Jul 26, 2014 at 09:29:47PM +0200, Philipp Kern wrote:
  Oh well, you just uncovered a bug that was not exposed widely because 
  there's
  a fallback P-a-s in the toplevel directory:
  
   * All the triggers source triggers/common.
   * common says at the top:
 PAS_BASE=/srv/buildd.debian.org/web/quinn-diff
 PAS_FILE=$PAS_BASE/$SUITE/Packages-arch-specific
   * $SUITE is set subsequently but the file has already been source and hence
 we get /srv/buildd.debian.org/web/quinn-diff//Packages-arch-specific 
  for
 all suites.
   * This file exists and points to the sid checkout.
 /srv/buildd.debian.org/web/quinn-diff/Packages-arch-specific - 
  sid/Packages-arch-specific
  
  I'll fix that. Thanks.
 
 should be fixed. Hopefully.

Thanks.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/1406494423.21696.0.ca...@jacala.jungle.funky-badger.org

P-a-s for stable and stable-security

[Adam D. Barratt]

Hi,

tl;dr: do stable and stable-security chroots apply P-a-s correctly?

DSA 2952-1 updated kfreebsd-9 in wheezy-security. As it was built on all
architectures for which kfreebsd-9 was available in wheezy, it was then
also accepted in to proposed-updates. It appears that packages for some
other architectures - arm{el,hf}, ia64, mips, powerpc, s390{,x} and
sparc - were subsequently built by the buildds in wheezy chroots.

The kfreebsd-9 source package in wheezy has Architecture: any all.
That changed in unstable at some point last year, and the package was
subsequently removed from the sid branch of P-a-s in May.

However, the wheezy branch of P-a-s still contains:

%kfreebsd-9: kfreebsd-i386 kfreebsd-amd64 i386 amd64 mipsel hurd-i386 # 
freebsd kernel 8.x

This raises a couple of questions:

- are the wheezy w-b databases filtered using the wheezy branch of
P-a-s?

- are the wheezy and wheezy-security w-b databases filtered using the
_same_ branch of P-a-s?

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/1405187709.24592.31.ca...@jacala.jungle.funky-badger.org

Re: backport of dpkg (= 1.17.2) and apt (= 0.9.16.1) for build profiles

[Adam D. Barratt]

On Sat, 2014-04-26 at 16:58 +0200, Philipp Kern wrote:
 [+ debian-admin, debian-wb-team ]
 
 On Sat, Apr 26, 2014 at 07:15:03AM +0200, Johannes Schauer wrote:
[...]
  12:33  Ganneff ftpmaster takes what dsa runs. dsa takes backports.
  12:35  adsb I'd suspect they'd prefer a stable update for that sort of 
  thing. but icb(e)w
  12:35  ansgar And support in w-b, apt (in stable), ... if there are 
  incompatible changes to Build-Depends.

fwiw it's generally considered polite to ask before quoting people's IRC
comments.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/1398528667.15557.24.ca...@jacala.jungle.funky-badger.org

Re: build profiles support of buildd infrastructure

[Adam D. Barratt]

On 2014-02-25 10:31, Adam D. Barratt wrote:

On 2014-02-25 9:55, Johannes Schauer wrote:
with dpkg 1.17.2 we now have support for build profiles in dpkg. Build 
profiles
allow to enable or disable dependencies and build or not build binary 
packages
when one or more build profiles are activated. For build dependencies 
this

looks like this:

Build-Depends: foo, bar-doc !profile.stage1

[...]
Before packages with above syntax can be uploaded, we have to make 
sure that

wanna-build, dak and sbuild support them.


The more general issue is that you need dpkg in stable to support them.

A lot of Debian infrastructure that touches source packages runs on
stable (for fairly obvious reasons) and expects things like
dpkg-source -x o just work[tm].


Replying to myself, I'm not sure extracting the package actually cares 
about the syntax of Build-Depends. Lots of other things will though, 
like pbuilder and lintian.


Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/317f9f6d75598bcb22b61204b3b6d...@mail.adsl.funky-badger.org

Re: build profiles support of buildd infrastructure

[Adam D. Barratt]

On 2014-02-25 10:49, Johannes Schauer wrote:

Quoting Adam D. Barratt (2014-02-25 11:35:37)

 The more general issue is that you need dpkg in stable to support them.

 A lot of Debian infrastructure that touches source packages runs on
 stable (for fairly obvious reasons) and expects things like
 dpkg-source -x o just work[tm].


Debian testing already includes dpkg ( 1.17.2) so dpkg should support 
the

required bits in jessie.


Indeed, and I see one of your later mails has now clarified that you're 
trying to get support ready for jessie. I'd (apparently incorrectly) 
assumed that you were trying to get the field supported /before/ the 
jessie release.


Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/a6399c0ab502d224f374956cab20e...@mail.adsl.funky-badger.org

Re: please give-back pyzmq on s390x

[Adam D. Barratt]

On Mon, 2014-01-13 at 20:19 +0100, Julian Taylor wrote:
 the underlying issue was in zeromq3 which is now fixed.

Given back.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1389644021.4536.17.ca...@jacala.jungle.funky-badger.org

Re: give-back ruby-kyotocabinet kfreebsd-amd64

[Adam D. Barratt]

On Mon, 2014-01-06 at 01:40 +0100, Christian Hofstaedtler wrote:
 please retry building of ruby-kyotocabinet on kfreebsd-amd64.
 I tried reproducing the build failure ('Terminated') on falla.d.o,
 but the package built fine there.

It looks like this was done; the package has been installed on
kfreebsd-amd64 for a couple of hours now.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1389037366.30047.15.ca...@jacala.jungle.funky-badger.org

Re: please give back beast-mcmc on i386

[Adam D. Barratt]

On 2013-11-04 15:12, Thorsten Alteholz wrote:

package beast-mcmc fails on i386 due to a missing dependency on
libhmsbeagle-java (= 1.1r1092).
As this package already migrated to testing, the dependency should be
fulfilled now.


It may well be, but there's now a different issue:

  beast-mcmc build-depends on missing:
  - libcolt-java

For reference, packages in such a state (BD-Uninstallable) will be 
automatically tried once all the build-dependencies are available (and 
installable), they don't need to be manually handled.


Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1eb84023dd8ef99dc6d6126be19df...@mail.adsl.funky-badger.org

Re: please give back beast-mcmc on i386

[Adam D. Barratt]

[not sure if you're subscribed, so erring on the assumption that you're
not; please say if you'd prefer not to get CCed]

On Mon, 2013-11-04 at 19:05 +0100, Thorsten Alteholz wrote:
 On Mon, 4 Nov 2013, Adam D. Barratt wrote:
  For reference, packages in such a state (BD-Uninstallable) will be 
  automatically tried once all the build-dependencies are available (and 
  installable), they don't need to be manually handled.
 
 Hmm, both libcolt-java from the colt source package [1] and 
 libhmsbeagle-java from the libhmsbeagle source package [2] migrated to 
 testing. Shouldn't that be enough to have the build-dependencies available 
 and installable? I guess I need another hint about the missing detail ...

Aha, I see what we've been missing:

libcolt-java |  1.2.0+dfsg-2 | unstable/non-free | all

The buildd network won't install non-free packages in order to build
other packages (yes, that's #719626).

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1383591667.24015.17.ca...@jacala.jungle.funky-badger.org

Re: Please give back csound on mipsel

[Adam D. Barratt]

On Fri, 2013-09-27 at 15:07 -0400, Felipe Sateler wrote:
 gb csound_1:6.00.1~dfsg-2 . mipsel .
 
 
 Please give csound back.

Done.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1380311289.22427.10.ca...@jacala.jungle.funky-badger.org

Re: haskell-tls-extra binnmu needed on powerpc

[Adam D. Barratt]

On Fri, 2013-09-20 at 00:20 -0400, Joey Hess wrote:
 nmu haskell-tls-extra_0.6.3-2 . powerpc . -m 
 'libghc-cryptohash-dev-0.8.4-ed942 not available any more'
 
 This is a retry of the binnmu since the previous one failed due to
 praetorius running out of disk space.

In which case, it needs a give-back, rather than a binNMU; done.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1379652175.5669.32.ca...@jacala.jungle.funky-badger.org

Re: Please let openttd 1.3.2 wait for grfcodec 6.0.3 on mips

[Adam D. Barratt]

On Wed, 2013-09-18 at 20:21 +0200, Matthijs Kooijman wrote:
 At the same time, a new version of openttd, which build-depends on
 grfcodec is available, which I've just uploaded to unstable. I'd like to
 be sure that building it will not be attempted using the broken grfcodec
 6.0.2 build on mips, hence this request.
 
 dw openttd_1.3.2-1 . mips mipsel . -m 'grcodec (= 6.0.3)'

Done, with the missing f. :-)

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1379532549.27643.6.ca...@jacala.jungle.funky-badger.org

Re: Please give-back mplayer2 on kfreebsd-i386

[Adam D. Barratt]

On Thu, 2013-09-05 at 20:25 +0200, Sebastian Ramacher wrote:
 mplayer2 failed to build on kfreebsd-i386 due to #720440 which is now
 fixed in mpg123. I confirmed on fischer that mplayer2 now builds
 successfully on kfreebsd-i386, so please give it back:
 
  gb mplayer2_2.0-701-gd4c5b7f-2 . kfreebsd-i386

Done.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1378413683.31389.7.ca...@jacala.jungle.funky-badger.org

Re: binNMU for schroot

[Adam D. Barratt]

On Tue, 2013-09-03 at 22:27 +0100, Roger Leigh wrote:
 Would it be possible to schedule schroot for a rebuild/binNMU in
 unstable, and experimental if possible.  I'd like it to be rebuilt
 against the new Boost 1.54 library packages.  If a maintainer upload
 is preferable, I could do that instead.

fwiw, debian-release is the standard contact point for binNMUs.

In this particular case, schroot/unstable was already rebuilt against
boost 1.54 within the past few days.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1378244372.28836.79.ca...@jacala.jungle.funky-badger.org

Re: KDE 4.10 givebacks

[Adam D. Barratt]

On Mon, 2013-07-22 at 15:02 +0200, Pino Toscano wrote:
 Apparently rocs still fails :/
 The file it is compiling (DotGrammar.cpp.o) is a C++ graphviz grammar 
 parser written in Boost.Spirit (so template metaprogramming), which 
 indeed makes it take quite some memory, possibly more than what rem 
 has...
 Would it be possible to give rocs back, trying to make it build on a 
 buildd with more RAM?

I've given it back again, but have no useful way of influencing where it
gets built.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1374602488.32280.7.ca...@jacala.jungle.funky-badger.org

Re: KDE 4.10 givebacks

[Adam D. Barratt]

On Tue, 2013-07-16 at 15:37 +0200, Pino Toscano wrote:
 * kdepimlibs/ia64
 ld failed with an assertion:
 gb kdepimlibs_4:4.10.5-1 . ia64

As per #debian-release, this has been done already.

* meinproc4 segfaults on mipsel
 it seems like on some builds (all done on rem) meinproc4 crashed; all 
 the other buildds did not show any issue, and even other builds on rem 
 were okay:
 gb blinken_4:4.10.5-1 . mipsel
 gb kanagram_4:4.10.5-1 . mipsel
 gb kbruch_4:4.10.5-1 . mipsel
 gb kcalc_4:4.10.5-1 . mipsel
 gb kgamma_4:4.10.5-1 . mipsel
 gb kgeography_4:4.10.5-1 . mipsel
 gb kmplot_4:4.10.5-1 . mipsel
 gb kolourpaint_4:4.10.5-1 . mipsel
 gb kstars_4:4.10.5-1 . mipsel
 gb ktimer_4:4.10.5-1 . mipsel
 gb kwallet_4:4.10.5-1 . mipsel
 gb rocs_4:4.10.5-1 . mipsel

All done.

 * kdepim/sparc
 the build died because of an ICE:
 gb kdepim_4:4.10.5-1 . sparc

This was already done; the package has been in the archive since Tuesday
night.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1374169794.14751.8.ca...@jacala.jungle.funky-badger.org

Re: Give-back oce on mipsel

[Adam D. Barratt]

On Tue, 2013-07-16 at 09:28 +0200, D. Barbier wrote:
 Could you please give-back oce on mipsel, eysler had been unable to
 setup the chroot.

Done.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1374169923.14751.9.ca...@jacala.jungle.funky-badger.org

Re: please retry libvigraimpex

[Adam D. Barratt]

On Mon, 2013-07-01 at 19:57 +0200, Andreas Metzler wrote:
 On 2013-06-21 Andreas Metzler ametz...@downhill.at.eu.org wrote:
  libvigraimpex FTBFS on some archs. The errors are unreproducible (I
  actually did successful builds on porter machines for s390x and
  mipsel) and the source identical previous upload to experimental
  (1.9.0+dfsg-3) built successfully about a month ago.
 
  Please retry the failed builds.
 
  gb libvigraimpex_1.9.0+dfsg-4 . mipsel s390 s390x
 
 Ping?

Done.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1372702107.23327.6.ca...@jacala.jungle.funky-badger.org

Re: Please give back k3d on several arches

[Adam D. Barratt]

On Mon, 2013-07-01 at 19:07 +0100, Manuel A. Fernandez Montecelo wrote:
 A recent upload of gtkmm2.4 caused some problems with k3d, please give
 back k3d in the arches where it was not built correctly due to this
 version being present (1:2.24.3-1).  Now a new upstream version fixing
 this problem (1:2.24.4-1) has built in all of the supported
 architectures, so the rebuild should get the package past this
 problem.
 
 gb k3d_0.8.0.2-20 . armel hurd-i386 i386 kfreebsd-amd64 kfreebsd-i386
 mips mipsel s390 s390x

Done.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1372702219.23327.7.ca...@jacala.jungle.funky-badger.org

Re: Please give back haskell-conduit on armel

[Adam D. Barratt]

On Tue, 2013-06-11 at 17:33 +0100, Colin Watson wrote:
 Now that GHCi is disabled on ARM, haskell-conduit should be buildable
 (because doctests will automatically be skipped).
 
 gb haskell-conduit_1.0.5.1-1 . armel

Given back.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1370975478.16430.4.ca...@jacala.jungle.funky-badger.org

Re: give backs for kfreebsd-i386

[Adam D. Barratt]

On Thu, 2013-06-06 at 16:08 +0200, Petr Salinger wrote:
 Please retry (due to #710841):
 
 kdenetwork
 mysql-workbench

Given back.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1370554667.15198.19.ca...@jacala.jungle.funky-badger.org

Re: Please, give back yade on ia64

[Adam D. Barratt]

On Wed, 2013-06-05 at 21:00 +0200, Anton Gladky wrote:
 yade 0.97.0-4 failed to build on ia64 with segmentation fault.
 Last 4 times yade failed on mundy. Other machines are building
 it fine. Please, give it beack:

Done.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1370554746.15198.20.ca...@jacala.jungle.funky-badger.org

Re: please give-back atlas on wheezy-proposed-updates until it is successfully built everywhere

[Adam D. Barratt]

On Tue, 2013-05-28 at 17:39 +0200, Andreas Beckmann wrote:
 (Is there another flag needed to denote the target wheezy? I don't see
 anything in http://release.debian.org/wanna-build.txt)

See the first item under notes.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1369766153.18388.4.ca...@jacala.jungle.funky-badger.org

Re: please give-back atlas on wheezy-proposed-updates until it is successfully built everywhere

[Adam D. Barratt]

On Tue, 2013-05-28 at 20:45 +0200, Andreas Beckmann wrote:
   The TARGET distribution is only needed if the request
  does not target unstable, i.e. experimental or wheezy.

At the risk of seeming picky, you mean e.g. here.

 Hmm, wheezy or stable ?

Either will work. The buildds use codenames, but the wb tool which the
commands are fed in to supports the codename / distribution alias list
exported by wanna-build.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1369766953.18388.7.ca...@jacala.jungle.funky-badger.org

Re: please give-back atlas on wheezy-proposed-updates until it is successfully built everywhere

[Adam D. Barratt]

On Tue, 2013-05-28 at 20:45 +0200, Andreas Beckmann wrote:
 The following might be more helpful:
 
   gb|dw|nmu PKGS_VER . ARCHES [ . TARGET ] [ . -m 'changelog entry/dep-wait 
 expr.' ]

fwiw, the reason I didn't use that approach initially was because it
makes the line  80 characters. That may not be an issue, of course.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1369767163.18388.9.ca...@jacala.jungle.funky-badger.org

Re: Please give back firebird2.5 to sparc

[Adam D. Barratt]

On Sun, 2013-03-24 at 15:05 -0400, Michael Gilbert wrote:
 Please give back firebird2.5 to sparc.  It had a compiler segfault
 during the build:
 https://buildd.debian.org/status/fetch.php?pkg=firebird2.5arch=sparcver=2.5.2~svn%2B54698.ds4-3stamp=1363963113

This was already handled via a request on IRC.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1364166236.24071.47.ca...@jacala.jungle.funky-badger.org

Re: Please giveback python-numpy/1:1.7.0~rc1-1 on s390x

[Adam D. Barratt]

On Tue, 2013-01-01 at 14:51 +0100, Sandro Tosi wrote:
 *** glibc detected *** /usr/bin/perl: malloc(): memory corruption
 (fast): 0x02443018 ***
 [8]
 dpkg-shlibdeps: error: dpkg-query --control-path libc6:s390x
 shlibs died from signal 6
 
 I think it's a transient error, or at least not related to the
 package, so can you please give it back?

It's a known phenomenon at least; given back.

fwiw, it helps to indicate when the package in question isn't the
unstable version, as the suite has to be explicitly passed to
wanna-build in such cases.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1357048803.15480.7.ca...@jacala.jungle.funky-badger.org

Re: mips build failures block testing migration

[Adam D. Barratt]

On Fri, 2012-12-28 at 17:28 +0100, Ivo De Decker wrote:
 Some build seem to be stuck on mips.

It looks like there's an issue with one of the buildds, indeed.

 For these, the missing builds block testing migration:
 
 qt4-x11
 lowmem (also Too young, only 5 of 10 days old)
 adplug (this one isn't unblocked yet)
 
 
 Can the builds be retried, or is there some other problem preventing this?

All given back.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1356720377.4281.2.ca...@jacala.jungle.funky-badger.org

Re: powerpc build of libatomic-ops in t-p-u

[Adam D. Barratt]

On 01.12.2012 11:10, Ivo De Decker wrote:
The t-p-u upload of libatomic-ops failed to build on powerpc. There 
is no

build log on

https://buildd.debian.org/status/package.php?p=libatomic-opssuite=wheezy

Can the build be tried again?


There's no log because it hit #680100, so trying it again isn't likely 
to make us very popular. :-) I've set the build to failed now with a 
reference to the bug so it should be clearer.


That bug's fixed in unstable; if someone could isolate the fix and 
apply it to the tpu...


Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/f15d1ba69475c4fc7b18566fc242e...@mail.adsl.funky-badger.org

Re: no change - why? Re: please rebuild mgltools-sff

[Adam D. Barratt]

On Sat, 2012-08-18 at 21:58 +0200, Wouter Verhelst wrote:
 On Sat, Aug 18, 2012 at 08:46:57PM +0200, Steffen Möller wrote:
  https://buildd.debian.org/status/package.php?p=mgltools-bhtree
  but all I get is
  https://buildd.debian.org/status/package.php?p=mgltools-sff
 
 Yes, I was actually expecting that.
 
 It doesn't say build-dep, it says BD-Uninstallable. That's an
 automated process which checks build-dependencies for installability.
 
 Why it's claiming that, though, I couldn't tell. A quick check does show
 that mgltools-bhtree is available (in non-free) for powerpc, yet when I
 add non-free to a sources.list file on a powerpc sid chroot, it doesn't
 even find it (though it does find the source).
 
 I'm probably missing something. Adding wb-team to Cc, maybe they've got
 a clue?

Unless something's changed quite recently, non-free buildd chroots don't
include non-free binary packages files so non-free sources that have
non-free build-dependencies won't get auto-built.

Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1345321076.31960.119.ca...@jacala.jungle.funky-badger.org

Re: powerpc: leptonlib stuck in Built state

[Adam D. Barratt]

On Sun, 2012-07-29 at 20:02 +0200, Hector Oron wrote:
 2012/7/29 Jonathan Nieder jrnie...@gmail.com:
 
  Based on [1], leptonlib was built on praetorius 8 days ago but
  never got uploaded.  Known problem?  Does it need any nudging
  to get put into place?
 
 JFYI, I am experiencing same problem with aircrack-ng:
   https://buildd.debian.org/status/package.php?p=aircrack-ng

No, that's a different problem, on a different buildd, assuming you were
referring to the powerpc build.

praetorius isn't auto-signed (unless something's changed very recently)
so needs manual intervention from the admin, i.e. signing the logs.  The
fact that leptonlib is in built implies that a signed log hasn't been
received.

aircrack-ng, otoh, was uploaded from poulenc but never marked as
installed.  That usually implies that either the package never reached
ftp-master or it was rejected there for some reason.

The queued log on ftp-master says:

Jul 28 18:55:23 processing /aircrack-ng_1.1-5_powerpc.changes
Jul 28 18:55:23 /aircrack-ng_1.1-5_powerpc.changes contained lines outside the 
pgp signed part, cannot process
Jul 28 18:55:23 Removing /aircrack-ng_1.1-5_powerpc.changes, but keeping its 
associated files for now.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1343586433.18013.125.ca...@jacala.jungle.funky-badger.org

Re: give back plastimatch

[Adam D. Barratt]

On Sat, 2012-07-28 at 19:01 +0200, Mathieu Malaterre wrote:
 Please give back plastimatch:
 
 gb plastimatch_1.5.11+dfsg0-1 . i386
 
 insighttoolkit 3.20.1+git20120521-3 was uploaded to fix an issue with
 a new behavior with gcc 4.7.0 and signaling NaN on i386 (#682805)

Done.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1343503983.18013.97.ca...@jacala.jungle.funky-badger.org

Re: Bug#681101: nmu: basic256_0.9.6.69a-1

[Adam D. Barratt]

Hi,

On 10.07.2012 17:09, Ryan Kavanagh wrote:

basic256 used to FTBFS on ia64[0] due to
/usr/lib/ia64-linux-gnu/libxml2.so.2: undefined reference to
`gzopen64@ZLIB_1.2.3.3'
I've successfully test-built it on merulo.debian.org. The ia64 buildd
should thus try building it again.

gb basic256_0.9.6.69a-1 . ia64


debian-release only handles binNMU requests; please direct other 
wanna-build requests to $arch@buildd.d.o or debian-wb-team@ldo (CCed)


Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/8a5a81eae6573aac2cf056086a152...@mail.adsl.funky-badger.org

Re: gb burp_1.3.8-1 . hurd-i386

[Adam D. Barratt]

On Fri, 2012-06-22 at 10:28 +, Bas van den Dikkenberg wrote:
   Thehurd-i386  version fail due to bug in gblicIt should be fixed by
 the latest glibc upload.
 
  
 
 gb burp_1.3.8-1 .hurd-i386

debian-release is not the correct contact point for that - you want
debian-wb-team or $arch@buildd.d.o ; copying the former.

Regards,

Adam



-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1340392553.27805.14.ca...@jacala.jungle.funky-badger.org

Re: gnucash 'BD-Uninstallable' on hurd-i386

[Adam D. Barratt]

On Wed, 2012-06-13 at 20:46 +0200, Micha Lenk wrote:
   gnucash (= 1:2.4.10-3) build-depends on one of:
   - libgoffice-0.8-dev (= 0.8.17-1.1+b1)
 
 But the package libgoffice-0.8-dev (= 0.8.17-1.1+b1) seems to be
 available in the archive since almost two months now, as could be seen
 here: https://buildd.debian.org/status/package.php?p=goffice
 
 So I would expect that gnucash should be in state 'Needs-Build' instead.
 Is there anything that I am missing?

{build-,}depends on one of means that the dependency is uninstallable;
if it's not available, edos-debcheck says {build-,}depends on missing
$pkg instead.

Digging a little deeper in to the packages files gives:

libgoffice-0.8-dev (= 0.8.17-1.1+b1): FAILED
  libgoffice-0.8-dev (= 0.8.17-1.1+b1) depends on one of:
  - libgoffice-0.8-8 (= 0.8.17-1.1+b1)
libgoffice-0.8-8 (= 0.8.17-1.1+b1): FAILED
  libgoffice-0.8-8 (= 0.8.17-1.1+b1) depends on missing:
  - libgoffice-0.8-8-common (= 0.8.17-1.1+b1)

That looks like a bug in the packaging, given that -common is arch:all
and therefore still at 0.8.17-1.1.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1339613882.10704.5.ca...@jacala.jungle.funky-badger.org

Re: gb feel++_0.91.3-4 . ia64

[Adam D. Barratt]

On Wed, 2012-04-18 at 07:14 +0200, Christophe Prud'homme wrote:
 This package failed on ia64 due to a lack of disk space on the
 builder.

Given back.

Regards,

Adam



-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1334858983.17116.0.ca...@jacala.jungle.funky-badger.org

Re: Bug#629952: subversion fails to build for stable(-security) on kfreebsd

[Adam D. Barratt]

On Sun, 2012-01-22 at 22:32 +0100, Aurelien Jarno wrote:
 On Thu, Jan 05, 2012 at 07:13:35PM +0100, Moritz Muehlenhoff wrote:
  On Thu, Jun 16, 2011 at 07:33:34PM +0200, Aurelien Jarno wrote:
   I think it builds fine on non-smp buildds, which at the time of the
   Squeeze release were fano and finzi. They have been switched to SMP a
   few weeks ago.
   
   Maybe we should switch them back to non-SMP? Alternatively I can build
   them manually to solve this situation, though it will only hide the
   problem.
  
  Can you upload manual builds? The archs need to be in sync for
  acceptance in the upcoming 6.0.4 point release of Squeeze.
 
 I have built both kfreebsd-amd64 and kfreebsd-i386 versions on
 non-SMP machines. I have just uploaded the resulting packages.

Thanks.  I can see the new packages on ftp-master, but they don't appear
to be on security.d.o yet?

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1327270300.6622.66.ca...@jacala.jungle.funky-badger.org

Re: Bug#629952: subversion fails to build for stable(-security) on kfreebsd

[Adam D. Barratt]

On Sun, 2012-01-22 at 23:20 +0100, Aurelien Jarno wrote:
 On Sun, Jan 22, 2012 at 10:11:40PM +, Adam D. Barratt wrote:
  On Sun, 2012-01-22 at 22:32 +0100, Aurelien Jarno wrote:
   I have built both kfreebsd-amd64 and kfreebsd-i386 versions on
   non-SMP machines. I have just uploaded the resulting packages.
  
  Thanks.  I can see the new packages on ftp-master, but they don't appear
  to be on security.d.o yet?
  
 Yes, I first uploaded them to ftp-master until I realized I should have
 uploaded them to security-master. I have done that half an hour ago,
 they arrived in the embargoed queue, and they should be publicly
 available in the next minutes.

Looks like they're there now; thanks.

Predictably the keys for most of the other architectures have expired in
the meantime, but that's another issue (that most of the current
recipients probably don't care about).

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1327271035.6622.68.ca...@jacala.jungle.funky-badger.org

Re: many builds missing for krb5-appl opensaml2 in stable-security

[Adam D. Barratt]

On Wed, 5 Oct 2011 12:43:26 +0200, Thijs Kinkhorst wrote:

Op maandag 03 oktober 2011 19:48:47 schreef Thijs Kinkhorst:

Hi Kurt,

Op zondag 25 september 2011 17:48:46 schreef Kurt Roeckx:
 On Sun, Sep 25, 2011 at 01:35:46PM +0200, Kurt Roeckx wrote:
  I also don't understand why they get rejected once the key
  expired.  The signatures are from before the key expired and
  perfectly valid.

 I should probably clarify this.  Of course you can't trust that
 the signature wasn't made after the key expired.

Why do these keys expire so quickly? Can't we have a really large 
window,
say a year, in which the key stays valid after the date it was last 
in

use?

 But we really should have a process so that once it's accepted it
 stays accepted, even when it moves to an other host.

How do you propose we solve this concrete issue?


I prefer that at least the krb5-appl and opensaml2 builds are made 
available
before the point release this weekend. Let me know what I can do to 
help.


As an additional data point, simply getting the missing builds uploaded 
won't currently resolve the issue for either krb5-appl or opensaml2, as 
some of the builds already in p-u-NEW are signed by now-expired keys.


Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/629c1d3e60da30dac0a0ba263393d...@adsl.funky-badger.org

Re: many builds missing for krb5-appl opensaml2 in stable-security

[Adam D. Barratt]

On Wed, 2011-10-05 at 18:49 +0200, Kurt Roeckx wrote:
 On Wed, Oct 05, 2011 at 12:15:40PM +0100, Adam D. Barratt wrote:
  
  I prefer that at least the krb5-appl and opensaml2 builds are made
  available
  before the point release this weekend. Let me know what I can do
  to help.
  
  As an additional data point, simply getting the missing builds
  uploaded won't currently resolve the issue for either krb5-appl or
  opensaml2, as some of the builds already in p-u-NEW are signed by
  now-expired keys.
 
 So now I got a bunch of REJECT mails, for the following
 packages:
 - krb5-appl
 - dovecot
 - php5
 - ejabberd

Yep.  They were rejected in order to be re-signed, at Thijs's request.
I did mention on #-buildd that the rejects would be occurring, but
apologies if I didn't do so in a sufficiently obvious way.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1317836340.4522.3.ca...@hathi.jungle.funky-badger.org

Re: Please gb fgrun against simgear-dev (2.0.0-4)

[Adam D. Barratt]

On Tue, 02 Aug 2011 08:49:07 +0100, Christopher Baines wrote:

The maintainer for simgear has been updating parts of the library for
the openscenegraph transition. However this appears to make fgrun
segfault (when built with simgear 2.0.0-2 at least).

  gb fgrun_1.5.2-1 . ALL . -m 'simgear-dev (= 2.0.0-4)'


There are a couple of issues with the above:

- give backs don't take a message argument; the above syntax would be 
for a dep-wait
- a give back or dep-wait only works when the package failed to build; 
in this case you need a bin-NMU


I've done this:

nmu fgrun . ALL . -m Rebuild against simgear-dev 2.0.0-4 to fix 
segaults

dw fgrun . ALL . -m 'simgear-dev (= 2.0.0-4)'

Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/fd14500b51a53763a537275ed1325...@adsl.funky-badger.org

Re: [experimental] Please requeue spark on powerpc

[Adam D. Barratt]

On Mon, 25 Jul 2011 10:58:14 +0200, Eugeniy Meshcheryakov wrote:

Details:
 * -d experimental to specify the target distribution (when it's not
   unstable).
 * -o because the package was marked as “Failed”, so needed an 
override,

   see below.
I was using information from 
http://release.debian.org/wanna-build.txt .
It does not mention -d and does not descrive -o flag. It could be 
useful

to add them there.


Actually, it *does* mention -o and even gives an example for 
give-backs.  Admittedly, it doesn't mention that this would also be 
required for packages currently marked as Failed, but it is there.


The -d flag isn't explicitly mentioned because it was assumed that 
whoever was performing the action would know it was required.  fwiw, the 
wb script also supports a syntax without the -d flag, namely:


  wb $action $pkgs . $arches [ . $dist ] [. $options ]

I'll look at updating the file on release.d.o, unless someone beats me 
to it.


Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/76da9d4a0906033e06df5a70593b0...@adsl.funky-badger.org

Re: Fwd: binNMUs for gdk-pixbuf upload

[Adam D. Barratt]

On Sat, 2011-07-16 at 10:25 +0200, Cyril Brulebois wrote:
 Forwarding to debian-release@, in charge of binNMUs.
 
 - Forwarded message from Michael Biebl bi...@debian.org -
[...]
 As part of the multiarch conversion I removed the libtool .la files.
 I checked all rdeps of gdk-pixbuf according to [1], and three library packages
 seem to be affected by that .la file removal, because they are listed in
 Build-Depends by other packages. Those packages are:
 
  libunicap2-dev (src:unicap)
  libvdk2-dev (src:vdk2)
  libmdc2-dev (src:xmedcon)
 
 Please schedule binNMUs for those packages with a dep-wait on  
 gdk-pixbuf_2.23.5-2

Scheduled, with a dep-wait on libgdk-pixbuf2.0-dev (= 2.23.5-2).

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1310810611.3606.4.ca...@hathi.jungle.funky-badger.org

Re: Quodlibet 2.3-1 vs. 2.2.99-1(+b1)

[Adam D. Barratt]

On Thu, 2 Jun 2011 13:09:43 +0200, Ondřej Kuzník wrote:

I'm helping maintain the package quodlibet and a really weird thing
has happened lately.

A transition has resulted in a binNMU on one of the binary packages
(quodlibet-ext). After that I packaged a new version of the software,
removing that binary package altogether and thus making quodlibet an
arch: all package. That one was later accepted in the archive but 
right

now packages.debian.org tells me that the last version of its binary
packages in 2.2.99-1(+b1) and that the last version of the source
package is 2.3-1 - the 2.3-1 binary packages have disappeared.


I think the problem is this:

 quodlibet |   2.2.99-1 |  unstable | source, all
 quodlibet |  2.3-1 |  unstable | source, all

which is most likely caused by quodlibet-ext still being in the archive 
in unstable.  You need to ask ftp-master to remove the old -ext 
packages, which will hopefully remove the 2.2.99-1 source and binary 
packages of quodlibet in the process; it might be worth checking with 
them that this will happen, just in case.


After a brief discussion on #debian-devel, KiBi suggested that I 
contact

your team.

Can you help me investigate what went wrong not to make the same 
mistake

again? What should I do next?


See above.  In any case, this is an issue on the archive side, not 
anything to do with the buildds.


Regards,

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/305223f6eacd306fe91aefbf2aee2...@adsl.funky-badger.org

Re: libforks-perl: gb libforks-perl_0.34-1 . armel

[Adam D. Barratt]

On Tue, 2010-09-28 at 21:30 +0100, Hector Oron wrote:
 I just build test on abel.d.o, same hardware as most buildds nowadays.
 
 [...]
 dpkg-deb: building package `libforks-perl' in
 `../libforks-perl_0.34-1_armel.deb'.
  dpkg-genchanges  ../libforks-perl_0.34-1_armel.changes
 dpkg-genchanges: including full source code in upload
  dpkg-source --after-build libforks-perl-0.34
 dpkg-buildpackage: full upload (original source is included)
 
 Could you give it back on armel? And close this bugreport if appropiate?

There's nothing to give-back - libforks-perl_0.34-1/armel has been in
the archive for three weeks now, having built successfully on alwyn.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1285710778.16049.184.ca...@hathi.jungle.funky-badger.org

Re: wanna-build state changes by dinstall

[Adam D. Barratt]

On Sun, 2010-06-06 at 13:12 +0200, Andreas Barth wrote:
 First, I intended to create a state known, which is the maximum of
 installed and needs-build, i.e. for take, failed, attempted, ... it
 behaves like needs-build, for binNMU it behaves like installed, and it
 doesn't set packages from needs-build to known unless directed to do
 so. Which of course means we need to adjust a couple of things, and we
 need an way to specify --needs-build. (Then I thought we could use
 use --binNMU 0 to say needs-build. That is an ugly hack, but gave me
 the following thought.)

fwiw, binNMU version 0 is already reserved, to mean cancel any current
binNMU and set the package back to Installed.

Adam


--
To UNSUBSCRIBE, email to debian-wb-team-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1275826004.29153.70.ca...@kaa.jungle.aubergine.my-net-space.net