Re: Bug#531581: Critical problems on hppa and ia64 buildds
Frank Küster wrote:
> Luk Claes <luk@debian.org> wrote:
>
>>> That doesn't solve my problem: Should I
>>> - make sure that the porters, buildd admins etc. are aware of the
>>> problem and simply close the bug?
>> You might want to downgrade the bug and only close it when it is realy
>> solved?
>
> And what would be the criterion for "solved"? After an analysis of
> N build logs of random packages on that buildd show no segfaults any
> more? I am not going to do that.
>
> It's not a bug in my package; I am not going to take responsibility for
> it.
Fine, though taking the trouble to talk to the porters might still be
worthwile.
>>>> The one on ia64 breaks the buildd's chroot and looks to be easily solved
>>>> by making sure that the maintainer scripts don't fail when the missing
>>>> command is not available.
>>> ?
>>>
>>> It could be easily solved by making sure that nothing on the buildd
>>> installs packages without installing their dependencies.
>> A patch is appreciated, thanks for your cooperation.
>
> Excuse me - that should already be guaranteed by dpkg, or am I missing
> something? If it isn't on that machine, what kind of patch should I
> write?
That it apparently is not already guaranteed by dpkg or it would not
happen and that it only seems to happen with your package?
Apparently you don't want to change your package to cope with it, so I
guess the only other option is to make the build environment cope with
your package (which might be actually implementing the behaviour you
expect and seems to be written in policy).
Cheers
Luk
Reply to: