Hello Andreas,Thank you for the update to Ogre 14. I'm so glad that it has finally reached testing, since I worked hard during months with the Games team to get it updated ! I think I forgot I had to update at some point on my main package...
Updating to 26.0.0 should be no harm indeed. I can't see in the diff why that would occur now. But if I look at the history on https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/sight.html, I can observe that it use to fail also on previous versions... So I take the hypothesis that this is not a new issue.
From the log you provided we can see all targets are not concerned. We can find the very first target (sight_core) and other high-end modules such as libsight_module_viz_scene3d. I can't really find a discriminant. All targets use the same linker flags. I don't see any fancy flag in our flags set.
We do generate files at build but they do not use timestamp or dates.My only guess so far would be the build order of the sources. We use file(GLOB_RECURSE... ) in our CMake scripts to list sources (which is actually explicitly said to be wrong by Kitware, but this is quite convenient...). This might give a different order depending on the system. At least I learned today it could... ;-)
I committed a patch, it builds on my workstation but I understand I can't make the repro test on my own, I would need two different configurations. So if I am right, would you mind to make a new upload so that we can check if this improves things on the "reproducible runners" ?
Best regards, *Dr. Flavien BRIDAULT* Director of Software Development IRCAD France & IRCAD Africa [email protected] <mailto:[email protected]> IRCAD France http://www.ircad.fr/ http://www.ircad.africa/ <http://www.ircad.fr/>Suivez l'IRCAD sur Facebook <http://www.facebook.com/pages/IRCAD/193785273990141>
*IRCAD France*Hôpitaux Universitaires - 1, place de l'Hôpital - 67091 Strasbourg Cedex - FRANCE
Le 09/06/2026 à 10:01, Andreas Tille a écrit :
Attention ! Ce message a ?t? envoy? depuis l'ext?rieur de votre organisation. Hi Flavien, I considered an easy task to solve s/libogre-1.12-dev/libogre-14-dev/ (bug #1138262) which would probably been possible even with the previous version. Unfortunately we did run into a reproducibly issue with version 26.0.0 and this became a thing now in Debian since it is requried that packages that once builded reproducibly have to keep this feauture. I admit I'm not sufficiently experienced - but here is the log claiming its not reproducible: https://eu-central-1.protection.sophos.com?d=reproduce.debian.net&u=aHR0cHM6Ly9yZXByb2R1Y2UuZGViaWFuLm5ldC9hbWQ2NC9hcGkvdjEvYnVpbGRzLzI2MDA3My9hcnRpZmFjdHMvNjY5NDUxL2RpZmZvc2NvcGU=&i=NjQxNDdjYjc3OWE3ZmQzMTQyZDg4YTcw&t=d0ZXVmFmazBnNnlwL2xXT21lSGxCL3dqMFhISUV2bWdzQ1h2b2MrbXdmdz0=&h=f65f5fc4deb2424da4ea974e3f71a4a3&s=AVNPUEhUT0NFTkNSWVBUSVZfvcCTmzaOZIcwaJNcq9k2EYwtp5TK0bdF-5_xAexoSw Can you have a look into this? Kind regards Andreas.
OpenPGP_signature.asc
Description: OpenPGP digital signature
