Re: Where did Bacula 1.38.11-7+b1 come from?
* John Goerzen:
> sid seems to contain bacula 1.38.11-7+b1, a binary-only NMU for i386,
> which breaks bacula-console and various other packages due to broken
> deps. The changelog file is signed only "buildd_i386-saens".
> packages.qa.debian.org doesn't know about 1.38.11-7+b1.
Here's the build log:
http://buildd.debian.org/fetch.cgi?&pkg=bacula&ver=1.38.11-7%2Bb1&arch=i386&stamp=1171847294&file=log
All official autobuilders are supposed to post their logs to
buildd.debian.org, AFAIK.
The problem seems to be that Architecture: all packages weren't
rebuilt. They refer to exact package versions in their dependencies,
which cannot be fulfilled by the binNMUed packages.
> I think someone deserves a serious thwacking... they obviously didn't
> even try to install the result of the NMU, filed no bug about it, etc.
Binary-only NMUs are a necessary evil. The implementation kind of
sucks, but I'm not sure how a better approach would look like. It's
not just the dependencies problem, it's also quite confusing that
you've got a source package which builds different binary package
versions on different architectures (but their are other ways to
achieve that). For instance, this pretty much rules out precise
tracking based on binary packages.
Reply to: