Bug#1137222: nmu: tetrinetx_1.13.16-22

Adrian Bunk <[email protected]> Thu, 21 May 2026 11:29:39 +0300
Newsgroups gmane.linux.debian.devel.bugs.general,gmane.linux.debian.devel.dpkg.general
Message-ID <ag7Cc0k1Y7GM8Pdm@localhost>
Control: reassign -1 dpkg
Control: retitle -1 Randomness in dependencies when symbols with different but equal versions are used

On Thu, May 21, 2026 at 09:40:49AM +0200, Jochen Sprickerhof wrote:
> Package: release.debian.org
> Severity: normal
> X-Debbugs-Cc: [email protected], [email protected]
> Control: affects -1 + src:tetrinetx
> User: [email protected]
> Usertags: armhf
> User: [email protected]
> Usertags: binnmu
> 
> Hi,
> 
> can you please trigger a binNMU of tetrinetx for reproduce.d.n:
> 
> nmu tetrinetx_1.13.16-22 . armhf . unstable . -m "Rebuild to fix Depends:"
> 
> We currently have this diff:
> 
> -Depends: libadns1t64 (>= 1.5.0~0), libc6 (>= 2.38)
> +Depends: libadns1t64 (>= 1.5.0~), libc6 (>= 2.38)
> 
> And I can't reproduce the buildd version with the 0 locally in a fresh
> sbuild chroot. I assume it was some bit flip on the buildds.

The same happened on hppa, which rules out a bit flip.

The root cause is in dpkg.

I do not understand the source of non-determinism in the output,
but symbols with different but equal versions are used by tetrinetx:

$ dpkg --compare-versions "1.5.0~0" "=" "1.5.0~"
$ echo $?
0
$

> Thanks!
> 
> Jochen

cu
Adrian