[Reproducible-builds] Recent issues with faketime in debian/rules during reproducible-builds builds

[Axel Beckert]

Hi,

Santiago Vila wrote:
> I can only give partial answers and minor comments, Holger and Mattia
> probably have an authoritative answer.
> 
> > Are not all chroots used by the reproducible-builds project setup
> > the same way?
> 
> Assuming that by "setup the same way" we are not talking about all the
> things that change between build1 and build2,

Yes, of course. :-)

> yes, the chroots should probably be setup "the same way" (or at
> least in a way that do not break packages gratuitously).

Ok.

> Also, we have such amount of things to fix and diffoscope logs to look
> at that for many things we rely on people telling us what's wrong, as
> in this case (btw: please consider joining us :-).

I've now joined the IRC channel and subscribed to the mailing list
(i.e. no more need to Cc me, but I won't mind if it will happen :-) as
it seems low traffic most of the time.

> > Does the project use pbuilder on Wheezy occassionally but not
> > always?
> 
> Holger and Mattia (root @ jenkins) can check this.

Mattia just answered that one on IRC: "all machines runs jessie and
all chroots are set up the very same way (changes between build1 and
build2 are done at run-time, the underlying chroot is the same)"

He also scheduled a new build of mp4h and it no more FTBFS'ed:
https://reproducible.debian.net/history/mp4h.html

So this was a non-deterministic issue as it seems.

		Regards, Axel
-- 
 ,''`.  |  Axel Beckert <abe at debian.org>, http://people.debian.org/~abe/
: :' :  |  Debian Developer, ftp.ch.debian.org Admin
`. `'   |  4096R: 2517 B724 C5F6 CA99 5329  6E61 2FF9 CD59 6126 16B5
  `-    |  1024D: F067 EA27 26B9 C3FC 1486  202E C09E 1D89 9593 0EDE