Bug#1144480: qgis regressed in reproducibility

Jochen Sprickerhof <[email protected]> Sat, 15 Aug 2026 20:19:25 +0200

Newsgroups	gmane.linux.debian.devel.bugs.rc
Organization	The Debian Project
Message-ID	<aoCtqvyJhrFR5imC__8360.54304834164$1786818076$gmane$org@mpd>

* Sebastiaan Couwenberg <[email protected]> [2026-08-15 19:50]:
>On 8/15/26 7:43 PM, Jochen Sprickerhof wrote:
>>src:qgis in unstable can't migrate to testing because the QA
>>infrastructure couldn't reproduce the some binary packages in
>>unstable while it could reproduce the version in testing [1, 2].
>
>AFAIK qgis has never been reproducible.

For completeness: Britney evaluates reproducibility per binary package 
independently. That's why it states:

Migration status for qgis (3.44.13+dfsg-1 to 3.44.13+dfsg-2): BLOCKED: Rejected/violates migration policy/introduces a regression
Issues preventing migration:
∙ ∙ Piuparts regression - https://piuparts.debian.org/sid/source/q/qgis.html
∙ ∙ Reproducibility regression on arm64: libqgis-app3.44.13, libqgis-gui3.44.13
∙ ∙ Reproducibility regression on armhf: libqgis-core3.44.13
∙ ∙ Reproducibility regression on i386: libqgis-app3.44.13
∙ ∙ Missing build on riscv64
∙ ∙ Lintian check waiting for test results on riscv64 - info
∙ ∙ Too young, only 3 of 5 days old
∙ ∙ Depends: qgis glibc
Additional info (not blocking):
∙ ∙ Updating qgis will fix bugs in testing: #1143800
∙ ∙ Not reproduced on amd64 (not a regression): python3-qgis-common

I.e. it is reproducible in testing except for python3-qgis-common.

signature.asc (application/pgp-signature, 833 B)