[Debian-med-packaging] Please add force hint to enable sra-sdk testing migration

[Emilio Pozuelo Monfort]

On 14/10/16 10:18, Andreas Tille wrote:
> On Thu, Oct 13, 2016 at 09:59:07AM +0200, Emilio Pozuelo Monfort wrote:
>> On 13/10/16 08:54, Andreas Tille wrote:
>>> Hi Julien,
>>>
>>> On Thu, Oct 13, 2016 at 12:14:48AM +0200, Julien Cristau wrote:
>>>>> please add a force-hint to add the testing migration of
>>>>>   
>>>>>      sra-sdk 2.7.0-1
>>>>>
>>>>> It was discussed with upstream that only amd64 architecture will
>>>>> be supported officially.  Since we try to fix build issues on other
>>>>> architectures step by step there is no point in delaying the
>>>>> testing migration for version 2.7.0-1 for amd64.
>>>>>   
>>>> NAK, this isn't how that works.  The package has out of date binaries in
>>>> unstable, that needs to be fixed one way or the other.
>>>
>>> Sorry for bothering you: what binaries are out of date, how does this
>>> come and how can I find this out?  Would a simple upload of a new
>>> package version fix this?
>>
>> sra-toolkit | 2.3.5-2+dfsg-1 | unstable        | i386, kfreebsd-amd64, kfreebsd-i386
>> sra-toolkit | 2.7.0-1        | unstable        | amd64
>>
>> https://buildd.debian.org/status/package.php?p=sra-sdk
>>
>> Either you fix the build (preferably), or you request the removal of the broken
>> binaries by filing a bug against ftp.debian.org. To fix the build, you need to fix
>>
>> https://buildd.debian.org/status/package.php?p=ncbi-vdb&suite=sid
>>
>> on at least i386.
> 
> Upstream explicitly confirmed that only amd64 is supported (for ncbi-vdb
> and sra-sdk).  I perfectly agree that it would be an optimal situation
> if also i386 would be supported but for the moment the unavailability
> for this architecture should not block the migration to testing.
> 
> So I filed #840683 to remove the unsupported architectures which was
> just closed thanks to quick ftpmasters. :-)
> 
> Am I correct now that a force-hint is apropriate now?

No, it should just migrate without one, given it's arch:any and not arch:all.

Cheers,
Emilio