Review: Syncing from testing a success?

Matt Zimmerman mdz at
Fri Apr 9 11:01:45 BST 2010

On Thu, Apr 08, 2010 at 09:41:48PM +0200, Martin Pitt wrote:
> In Lucid we have switched the default syncing source from Debian
> unstable to testing [1], to avoid getting only pieces of large transitions
> which we then need to untangle ourselves, and generally avoid
> importing major regressions which are held back from Debian testing
> through reported RC bugs.
> This comes at the cost of a bigger wave of new package versions in the
> next release, and with the increased cost of more manual sync requests
> from unstable.
> Usually we'd switch back to unstable for 10.10 (maverick), but now is
> a good time for a review how the syncing from testing actually worked
> in practice and whether we'll keep it for LTSes, all releases, or drop
> it again.

Thanks for initiating this review.

> So, in your opinion, did syncing from testing
>  (1) help to avoid introducing larger breakage into Ubuntu (for the
>      domain you are usually watching)

I found Lucid to be pretty well-behaved during active development, compared
to other cycles.

>  (2) meant a smaller or larger amount of review and sync requests

Can't we calculate this from data in Launchpad, rather than asking for
multiple opinions? :-)

 - mdz

More information about the ubuntu-devel mailing list