Empty build-id to make package reproducible

Otto Kekäläinen otto at debian.org
Sat Sep 1 22:15:57 BST 2018


la 1. syysk. 2018 klo 20.42 Mattia Rizzolo (mattia at debian.org) kirjoitti:
> The thing I can say is that that commit is not fixing *anything*.  The
> pipeline failed before and after
> https://salsa.debian.org/mariadb-team/galera-3/commit/1460cfa128fb457b5b5c60fcc5cac6faf5a216d5
> (as can be seen in that page quite clearly).  Instead, what fixed the
> pipeline was the next commit, which made DEB_BUILD_OPTIONS=nocheck
> actually work.
> Indeed, what was failing in the reprotest stage of the pipeline were the
> tests of the package, so disabling that made the pipeline pass.

First we fixed what r-b infra showed in diffoscope as the culprit,
then we fixed what reprotest showed as the failure on Salsa-CI. None
of the fixes were uploaded and now all of a sudden r-b pass..

> As a datapoint, the package is reproducible in our infra.

Apparently it now is..
2018-08-22 23:50 25.3.23-1 unstable amd64 reproducible

.. but it was not in unstable at the time we were investigating this
and trying to fix it:
2018-07-05 21:09 25.3.23-1 unstable amd64 unreproducible unreproducible

(from https://tests.reproducible-builds.org/debian/history/galera-3.html)

It seems we were trying to fix in the package an issue that was after
all in the r-b infra.

> > > then again, the debugging symbols for galera-3 look like they're being
> > > generated in a way that is pretty out-of-date, and hasn't been touched
> > > in at least 3 years, so maybe the maintainers don't care about these
> > > symbols very much:
> > >
> > >    https://salsa.debian.org/mariadb-team/galera-3/blame/master/debian/rules#L51
> >
> > This package is actively maintained and everything should be up to
> > date. If you are an expert on debug package rules stanzas, we are
> > happy to take any suggestions (or merge requests on Salsa) to make
> > that section not so "pretty out-of-date".
> I see dkg submitted a change, https://salsa.debian.org/mariadb-team/galera-3/commit/853d8dc20dbf67e9d8ac76668098a6b3560ab786
> however, I wonder if that actually works with --build-id=none, given
> that the debug symbols are placed next to the build-id?  (not sure,
> maybe there is a fallback, I haven't actually checked).

Well, as the package is now reproducible on r-b infra without any
changes, I'll revert these experiments we did that are in git but
never uploaded.

- Otto

More information about the Reproducible-builds mailing list