mksh in experimental (was Re: Builds taking far too long)
On Sun, 24 Jul 2016, Christian Seiler wrote:
> @Thorsten: because of the dietlibc security upgrade, I requested
> binNMUs of all affected packages (including mksh), but because
mksh in sid or older wasn’t affected, it only shipped the klibc
binaries, not the dietlibc binaries (and there was no dietlibc
arch without klibc, TTBOMK). The Built-Using header would have
shown that, anyway.
> your naming scheme for the experimental version the binNMU in
> unstable now supersedes experimental. (I didn't think of that
Oh, wow. Never saw that coming. I thought everyone just added
exp1… must have been +exp1 that got added…
> before I requested the binNMUs.) Could you re-upload the
> experimental package with a different version number (for example
> 52c-2+exp3) that doesn't clash with binNMUs, so that the security
> issue gets fixed also in experimental? Thanks!
Hmm, to be honest, I’ve got some work to do with mksh, to get
more bugfixes culminating in a new release done, and I don’t
really see the point – the experimental upload has done what
it was meant for, and it can just be superseded, or removed.
I could try working on the necessary fixes to what’s currently
in mksh CVS HEAD and then uploading it as snapshot to sid to
supersede the experimental version more quickly (within the
week, unless the heat continues as bad as last week), that
should help. *moves his TODO around*
bye,
//mirabilos
--
>> Why don't you use JavaScript? I also don't like enabling JavaScript in
> Because I use lynx as browser.
+1
-- Octavio Alvarez, me and ⡍⠁⠗⠊⠕ (Mario Lang) on debian-devel
Reply to: