Hi, Excerpts from Luke Kanies's message of Thu Sep 23 21:10:41 -0400 2010: > On Sep 23, 2010, at 9:23, James Turnbull <[email protected]> wrote: > > >>>> Why are there new features pushed in point releases? > >>>> > >>> One is a customer requested feature and the second is a "feature" in > >>> name only - the current pkgutil provider is broken. > > > > Customers who pay for support get priority tasking on features - our > > current model is into the next shipped release where possible. This may > > change in the future as it presents some challenges. > > This has largely always been our practice, but it has not come up all > that often, and it's only ever applied to berry simple features such > as the one mentioned. > > We've been wondering internally how long we'll get away with it, so > let us know if you think that ship has sailed. >
>From my POV I'd like to be able to push new micro-releases to Ubuntu stable releases. Doing so would require a Micro Release Exception [1]. [1]: https://wiki.ubuntu.com/StableReleaseUpdates/MicroReleaseExceptions In order to get such an exception the following criteria need to be met: 1. upstream supports micro-version updates to stable releases 2. upstream has a sufficiently high level of regression testing for their stable releases 3. regression tests are enabled in the package's build New features (even small) definitely don't comply with #1. Bug-fix-only patches should be pushed to a micro release. Having a document outlining what is considered a bug fix and what can be accepted in micro-releases would be helpful here. As for point #2 are there any metrics that shows code coverage of the regression tests? What's the status of the unit test (from test/) and spec tests (from spec/)? How often are regression tests run? Is there a CI system setup somewhere? #3 falls into the packaging domain. Currently unit and spec tests are available in the puppet-testsuite package. Thus tests can be run against the actual installed packages. IIRC there are 3-4 unit tests failing and 7-8 spec tests failing. I need to look into the failure - most of them are related to the fact that tests are not run from the source directory (some changes were also required to the test code in order to support running the tests from /usr/share/puppet/). This should probably fixed in the upstream code. -- Mathias Gug Ubuntu Developer http://www.ubuntu.com -- You received this message because you are subscribed to the Google Groups "Puppet Developers" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/puppet-dev?hl=en.
