Hi!
with a little bit of a delay I'd like to get the ball rolling on
Bigtop 0.6.0. Given our quarterly release cadence it
will come out at the end of March (Fool's day release
perhaps? ;-)).
The first order of business is for us to settle on the BOM
and supported platform list. The following JIRA is supposed
to track those things:
https://issues.apache.org/jira/browse/BIGTOP-816
I've put my stake in the ground in the comment section
of the JIRA, but please do comment and let me know what
do you think. Once we settle on a list that looks reasonable
to us I'd like to broadcast the decision to our newly minted
annou...@bigtop.apache.org so that the community at
large has a chance of chiming in.
Given our strong push into the bleeding edge Linux distros
the following 2 JIRAs seems like a must for the release:
1.1. Support next-gen service management frameworks
https://issues.apache.org/jira/browse/BIGTOP-103
https://issues.apache.org/jira/browse/BIGTOP-116
1.2. Refactor/unify init.d scripts
https://issues.apache.org/jira/browse/BIGTOP-22
I don't think we have much chance of working seamlessly
with Fedora/OpenSUSE if we don't commit to them.
On top of that, I would also like to propose that Bigtop 0.6.0
doubles down on the developer-centric and integration testing
aspects of our charter. To that end I'd like to nominate the
following projects to be defining for the release:
* Fully integrate Whirr with Bigtop puppet
* Have a trunk-driven upstream integration testing in place.
We would produce builds with one component take out of
trunk and the rest from the stable base of the last release of
Bigtop and then run integration tests on that combination. I'm
already engaged with HBase community on this one:
http://www.mail-archive.com/dev@bigtop.apache.org/msg01108.html
* Start tackling the issue of a flexible framework that tests
can utilize to do more cluster manipulation (ChaosMonket, etc.).
Thoughts?
Thanks,
Roman.
