On 09/24/2009 12:00 PM, Thomas S Hatch wrote:
> Its on the list, problem is the list is very long :)


Indeed.    This is a key issue.    As other projects come up, my time to 
devote to Cobbler is not 100%.

As I'm mostly one person, I don't scale -- Lately there has been good 
interest in feature patches, but not a lot of interest in bugfix 
patches.   We've got to fix this and make this better.    This is very 
important for the project.

Thus, in order to get 2.0 pushed out from testing to stable more 
quickly, we need to grow the community more and get folks to be able to 
pitch in and fix things.  This was part of the idea around "Learn to 
Hack on Cobbler Week", which, admittedly, didn't get many bites.

Help in submitting patches is the best thing we can do to get Cobbler 
2.0 ready-to-roll into production sooner -- short term.   Long term, 
there is more we can do.   While 2.0 was a refactoring/core-cleanup 
release, 2.2 will probably have a lot of focus on test improvements 
(they can also be merged to and be run against 2.0).

Another thing we can do is establish a much more comprehensive 
functional test suite -- which, I think, for me, is the #1 feature I 
want to get going for 2.2.   This would involve end-to-end imports, 
creation of VMs (image and kickstart based), validation of DHCP and DNS 
modules, etc, etc.    This could be a *quite* sizable effort but could 
help us a good bit.   On the downside, it would require that new 
features coming in also came with tests as a criteria for patch 
acceptance.    (Web application testing, due to time constraints, I 
expect would be deferred for later -- but if someone wants to try to get 
something together using selenium (or equivalent) that could be very 
interesting.

Thoughts and volunteers?

--Michael


_______________________________________________
cobbler mailing list
[email protected]
https://fedorahosted.org/mailman/listinfo/cobbler

Reply via email to