Control: severity -1 serious Andreas Barth <a...@ayous.org> writes: > checking with http://release.debian.org/wheezy/rc_policy.txt I still > don't see why this bug is serious or above. Please note that the > mentioned document is the canoncial definition of release critical > bugs.
FHS violations are mentioned there. Also note that other uses of site-specific directories such as /srv or /opt even get an automatic rejection from dak (if lintian can detect them). Having a system user create a home directory in /home also makes the package uninstallable on systems with /home on nfs (or other filesystems where root does not have write permissions). While looking at it, I also noticed that the two binary packages create the same system user with different options: dsc-statistics-collector.postinst: adduser [...] --disabled-password [...] dsc-statistics-presenter.postinst: adduser [...] --disabled-login [...] So what you end with depends on the order you install the packages which it probably not a good thing. > Setting the bug to important only means I don't think it is serious or > above. Otherwise, as always it is the maintainers decision to set the > bug to an appropriate severity (and of course, the maintainer is free > to set the bug to serious again, if he thinks the package is unfit for > release). Anyone disagreeing with the maintainer may try to convince > him, search for other people convincing the maintainer, or escalate > the topic to the tech ctte - as always. It does not seem to be that the maintainer disagrees with the "serious" severity after Andreas Beckmann's more detailed explanation. There aren't many examples of escalation to tech-ctte happening (a good thing), but the one I could find[1] ended with you redirecting it from tech-ctte to the release team. Did anything change since then? [1] <http://bugs.debian.org/582423#10> Ansgar -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
