Custom Query (1145 matches)

Reported by jdreed, 13 years ago.

The picture of Killian is from Flickr ( http://www.flickr.com/photos/22240293@N05/3930252680/), available under CC-BY-2.0. We attribute it in the copyright file, but it's unclear if that's good enough to satisfy the terms of the license? We could trivially add the author's name with GIMP or something.

(If someone can authoritatively say it's good enough in the copyright file, then close this ticket).

Reported by broder, 15 years ago.

Right now packages built by the autodebathenifier are going into proposed, and require a manual move by a maintainer to move into production. In theory, this is to prevent a successful build of a mysteriously broken package from screwing everything.

I don't think this makes sense. I think we should be moving autodebathenified packages (and especially openafs modules) immediately into production. We don't perform any testing of autodebathenified packages before making the move. Nor do we have any automated infrastructure that does it for us. Nor do we seem to have made any progress at all in writing such an infrastructure.

And I'm not even clear what sort of mysterious bug we're concerned with. To the autodebathenifier has been running as part of the Athena 10/Debathena infrastructure for well over a year now, and it's never failed in a way that caused it to upload a broken package.

All this extra step is serving to do is create more work for debathena-root without any apparent gain, and it causes additional delays for packages that users may be expecting. Therefore, I'm going to plan to reconfigure the autodebathenificator to upload directly to production in a week.

Reported by jdreed, 13 years ago.

If building for, say, natty, fails, it should not affect the other 7 releases we're building for. Autodebathenify should have per-release (possibly also per-package) .suppress files.

Of course, now that #243 is fixed, it may be easier to just stop using autodebathenify once the non-dkms releases fall out of support.