[pkg-bacula-devel] Reproducible build weirdness

[Sven Hartge]

Hi!

I've been looking into why reprotest fails.

As of current HEAD, only two main DEBs (and their dbgsyms) are different:

bacula-common_9.2.2-1_amd64.deb
bacula-director_9.2.2-1_amd64.deb

And inside those DEBs the following binaries differ:

/usr/lib/bacula/bdirjson
/usr/sbin/bacula-dir

*And* inside those binaries there is a strange path leakage, causing an
offset later on.

Somehow parts of the buildpath to smartall.h and mem_pool.h end up in
the binaries and *only* those binaries, which is doubly strange, because
smartall.h and mem_pool.h are included via lib/lib.h which is included
via lib/bacula.h which in turn is included in everything.

I've attached the logfile from reprotest with diffoscope in case you
want to take a look yourself.

Grüße,
Sven,
-------------- next part --------------
A non-text attachment was scrubbed...
Name: reprotest-1542058294.log.gz
Type: application/gzip
Size: 1052641 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/pkg-bacula-devel/attachments/20181112/9407e786/attachment-0001.gz>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://alioth-lists.debian.net/pipermail/pkg-bacula-devel/attachments/20181112/9407e786/attachment-0001.sig>