Sounds good, this is consistent with what the TC discussed. Thanks Jeremy! On 4/20/2020 3:58 PM, Jeremy Stanley wrote:
The Infrastructure team, and the CI team before it, has traditionally existed to care for the continuous integration and collaboration infrastructure on which the OpenStack community relies. With the rise of the OpenDev Collaboratory as a distinct effort outside of (but still primarily in service of) OpenStack, the majority of the team's former systems administration activities are no longer occurring under the authority of OpenStack. Most of the software and configuration management repositories previously in the care of the Infra team are also no longer official OpenStack deliverables, and have become part of OpenDev as well (and their biggest development effort, Zuul+Nodepool, was already spun out as an independent Open Infrastructure Project last year).
With the above responsibilities moved elsewhere, the existence of a formal team is less of a necessity. What remains is a need to support OpenStack's project-specific testing and collaboration tooling and services, primarily job configuration and other things which shouldn't currently be generalized into components of the OpenDev Collaboratory. To this end, I propose the creation of a new Testing and Collaboration Tools (TaCT) SIG which would serve the role currently occupied by the OpenStack Infrastructure team. I'm open to alternative names, but employing the vague term "infrastructure" has perpetually confused newcomers to our community, so this seems like an opportunity for something less overloaded.
The TaCT SIG would consist of the usual suspects: people who review OpenStack job configuration changes, people who dig into problems with test frameworks to unblock the integrated gate queue, people who figure out strange Python packaging related issues, people who help work out lapsed control of Launchpad admin groups... also, ideally, the person selected by the TC to serve as OpenStack's representative on the OpenDev Advisory Council would be heavily involved. Many of these activities are closely related to work the Quality Assurance team is doing, so hopefully folks who are active in QA will also participate in this SIG.
What do you think? Does this satisfactorily capture the infrastructure needs of the OpenStack community, and could it serve as an alternative to the current Infrastructure team?