On Wed, Aug 20, 2014 at 7:19 PM, Benedict Elliott Smith < [email protected]> wrote:
> The "run versions >= x.x.6" is IMO an out of date trope. Not only does 2.1 > have more than twice as many engineers working full time on it, and five > times the QA engineers (which is cumulative with prior QA), we are also > seeing many more users in the wild running release candidates and providing > valuable feedback before release. This is evidenced by the fact there have > been six release candidates, instead of just two for 2.0. > While I agree that Datastax is recently dedicating meaningful resources to the QA side of Cassandra, I do not feel that it is controversial to say that the QA of Cassandra as it relates to production operability has historically been lacking. I feel much the same way about Cassandra QA that Gandhi apocryphally felt about Western Civilization... I think it would be a good idea. Based on user reports and associated JIRA, the 2.0 series has been the least stable series of Cassandra since 0.8 or so. It's wonderful that a pre-release version has had slightly more testing than previous versions, but IMO the proof is in the pudding. When there is an actual released series of Cassandra in which I can recommend running a version under x.y.6 [1], I will be sure to stop linking that "trope." =Rob [1] (x.y.~8 for 2.0...)
