Review: Syncing from testing a success?

Michael Bienia michael at bienia.de
Fri Apr 9 12:49:58 BST 2010


On 2010-04-09 11:03:21 +0100, Matt Zimmerman wrote:
> On Thu, Apr 08, 2010 at 11:00:00PM +0200, Michael Bienia wrote:
> > On 2010-04-08 21:41:48 +0200, Martin Pitt wrote:
> > >  (5) anything else that caused or eased problems that you can think
> > >      of
> > 
> > Some packages might got no updates for lucid as the package didn't reach
> > testing (for whatever reasons) but we got it from syncing from unstable
> > in karmic. These packages need to get synced manually from unstable to
> > benefit from package updates.
> 
> Why?  Is it necessarily a bad thing if a package doesn't get updated?  If it
> doesn't enter testing, there may be a good reason for this (such as RC
> bugs).

No, it's not a bad thing. The "problem" is that we synced that package
from unstable in the past. And for whatever reason a newer version
doesn't make it into testing. So we are stuck with this potentially
buggy package till it reaches testing.
As I see it we have two options: a) continue syncing this package (for
now) from unstable and hope the problems why it doesn't migrate to
testing don't affect Ubuntu or b) remove the binary from Ubuntu till it
reaches testing. But currently we do neither (unless someone cares or
stumbles over this package).

This problems gets less when we continue to sync from testing as I
assume that most packages migrated to testing in between and we can stay
in sync with the version in testing. But when we sync from unstable
again, this problem will come back with the next switch to "sync from
testing" as we might got several packages synced from unstable that
won't migrate to testing for some time and the version we have is newer
than the one in testing but not kept uptodate with the version in
unstable.

Michael



More information about the ubuntu-devel mailing list