Nilesh Patra, on 2021-03-19 23:19:40 +0530: > On Fri, 19 Mar 2021 at 23:07, Étienne Mollier <firstname.lastname@example.org> > wrote: > > Nilesh Patra, on 2021-03-19 21:15:35 +0530: > > > radiant upgrade from buster seems to have a broken symlink as reported > > > in #985523. I tried fixing this with hints from previous scripts, and > > > taking a look at dpkg's maintscripts manpage. > > > > > > However, I'm not sure if these fixes are right. Please consider to > > > review and upload, or give me a green signal so I do so. > > > My commits are pushed to salsa > > > > I had a look, and apparently, the bug is caused by a case > > mismatch between the target of the link and it's actual file > > name: > > Could you confirm once if $sudo piuparts -a -d buster -d bullseye -d sid > radiant --existing-chroot=/srv/chroot/buster-amd64-sbuild > is the right way to reproduce the bug? This one goes OK for me. Already discussed rapidly on the chat, but for people following on the mailing list: this particular test is different from the previous bug reports. The key option to reproduce this is the flag "--fail-on-broken-symlinks", which is not set by default: $ sudo piuparts \ --fail-on-broken-symlinks \ --warn-on-leftovers-after-purge \ radiant_2.7.1+dfsg-3_amd64.changes If that helps, I noticed Andreas attaches the full piuparts log to the bug reports, which includes the full piuparts command line called (which I must admit has additionally a lot of various cruft, so I agree it may not the be the easiest to read.) Cheers, :) -- Étienne Mollier <email@example.com> Fingerprint: 8f91 b227 c7d6 f2b1 948c 8236 793c f67e 8f0d 11da Sent from /dev/pts/2, please excuse my verbosity.
Description: PGP signature