pkgsrc-Bulk archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: pkgsrc-current-destdir Linux 2.6.24-1-686-bigmem/i686 bulk build results 20080829.1720



Aleksey Cheusov <cheusov%tut.by@localhost> writes:

>  >>  >> Suppose
>  >>  >>   1) you need a package A
>  >>  >>   2) package A depends on B
>  >>  >>   3) B fails => A fails "indirectly"
>  >>  >> In such situtations (in most cases), it makes sense to contact
>  >>  >> maintainer of the package B. This is why your email is there.
>  >>
>  >>> No. In any case first person to be responsible for failure of package
>  >>> is that package maintainer.
>
>  >> In most cases dependancy is correct and the real problem is in
>  >> dependant package. This is why I include a maintainer of the dependant
>  >> package by default. In general only human can recognize the "right
>  >> person" to contact. Sometimes bugs in mk/ scripts or buildlinks3.mk or
>  >> anywhere else may cause the problem.
>
>> Don't talk about "most cases", there're correct ways and incorrect ones.
>
> It is trivially to proof that in most cases maintainer of failed
> dependancy is closer to the problem than that of "indirectly" broken
> package.

If it is trivially, prove it applied to wip/slate.

I don't care about "most cases", there are correct and incorrect
ways to handle specific problem. I provided evidence, why your
way is incorrect one.

> wip/slate and infrastructure failures are rare excludions.

Oh, really? Prove it.

I doubt you're even close to truth, since you have shown your
inability to handle this particular problem in correct way.

> I thought about inclusion of both maintainers. But I didn't do so
> because in this case I had problems with 90-columns limit of textual
> report.txt. I'll think about how to reorganize this section.

Then put your name right before maintainers, since you're even
closer to "problem detection" with your distbb. You're the
maintainer of the latter, aren't you?

> P.S.
> distbb provides separate file that contains only "directly" failed packages
> and many other useful META/*.txt files.
>
>    $ wget 
> http://mova.org/~cheusov/pub/pkgsrc-distbb/Linux/current/log/20080825.0717/META/packages_failed.txt
>    $ grep asau@ packages_failed.txt

Now we know another bug in your software: inconsistency of what is sent by mail
and what is saved in other files.

I don't care of that list on your server, since information in mail message
should be enough for me.


-- 
HE CE3OH...


Home | Main Index | Thread Index | Old Index