Sorry for the miss last week; it being a holiday in the US meant I was on the road managing tiny humans and a puppy with my partner and I failed to hand off update email responsibility to someone else. Which means we have three weeks to cover!
[New contributor Getting Started] As a new contributor we recommend starting in one of two places: Failing tests, or starter tickets we label "lhf" (low hanging fruit). Failing tests: https://issues.apache.org/jira/secure/RapidBoard.jspa?rapidView=496&quickFilter=2252 Unassigned starter tickets: https://issues.apache.org/jira/secure/RapidBoard.jspa?rapidView=484&quickFilter=2162&quickFilter=2160 Holding steady at 22 failing ticket JIRA's in the past few weeks. More on CI trends below with the new Butler tool below. For unassigned lhf, we're up from 11 to 12 on 4.0.2 (our next minor release) and holding steady at 14 on 4.1.0 (our next major release). We have a selection of people who have volunteered to be mentors; feel free to reach out to any of us if you have any questions on where to get started or just self-select from either of these above lists (list of mentor names can easily be seen here for now: https://issues.apache.org/jira/browse/INFRA-22556). [Dev list conversations] https://lists.apache.org/list?dev@cassandra.apache.org:lte=3w: In the past 3 weeks we've had some pretty interesting conversations surface and evolve. First off and quite exciting: Butler was introduced by a collection of contributors which helps us both see the current state of our CI across multiple branches as well as look more deeply into the failure rates of individual tests. The url to access this system is here: https://butler.cassandra.apache.org/#/ CEP-3 for Guardrails discussion closed up, was voted on, and passed the vote. https://cwiki.apache.org/confluence/display/CASSANDRA/CEP-3%3A+Guardrails, JIRA ticket: https://issues.apache.org/jira/browse/CASSANDRA-17146. Really excited to see how this work evolves. CEP-17 for the SSTable format API was also voted on and the vote passed. https://cwiki.apache.org/confluence/display/CASSANDRA/CEP-17%3A+SSTable+format+API, JIRA ticket: https://issues.apache.org/jira/browse/CASSANDRA-17056. Curious to see what various approaches to SSTables and persistence this eventually enables. We had a good discussion on mentoring and how to introduce new members to the project - that thread can be found here: https://lists.apache.org/thread/xr6b14w0b5sg00s79j4xlq6hls6fqs49. A lot of folks came forward volunteering to mentor new contributors to the project and we're working on making that list visible on the site, project, etc. Expect to see more on this front as the infra changes take place. A lively discussion is still going on around encryption at rest of SSTables within Cassandra. https://lists.apache.org/thread/6z1hkygqj48241sbgq8ogovkx5w7vv8p. There were multiple calls to the question of whether this ought to become its own CEP since it's a complex topic with a lot of tradeoffs. Benjamin hit the list talking about some of the pain points he and some other contributors have seen regarding inserting into collections on contended partitions ( https://lists.apache.org/thread/f3dl7rfc2kv9f5r9pxzyz6zojsss81b9). There's some work that's been put together by Benedict back in early 2020 that makes a substantial improvement in this area (warning: understatement of the week) and the question's up as to whether or not we want to include this in 4.0 or in 4.1. There's no clear right or wrong here; would definitely like to see more opinions on this relatively new thread. We have an ongoing discussion about YAML config, nesting vs. flat, how other projects do it, you name it. Another one of those "no obvious right answer" kind of situations; if you've been holding off on the topic thus far and have an opinion informed by experience, please chime in! https://lists.apache.org/thread/3l6f8lypoj5pj2bl3m9x4x75rsjchv4q And last but not least (non-inclusive; busy three weeks): Benjamin put out the call for a "get new contributors involved" advent calendar of tickets in December ( https://lists.apache.org/thread/3l6f8lypoj5pj2bl3m9x4x75rsjchv4q). While our LHF pool is pretty loosely curated at this point, there's a lot of potential here so if you can lend a hand please reach out to Benjamin. [CI Trends] New and flashy in this issue: what's going on with our CI? Well, a lot: https://butler.cassandra.apache.org/#/ First off, we actually _have_ butler now, which not only shows us trends for each branch but also allows you to see detailed results of which tests are consistently failing, which flaking, etc. While trunk is in a bit better of a state now than early November, 4.0 saw a pretty drastic jump to 45+ failures about a week ago. Builds 294 and 295 look to be having something systemic going on with them w/a variety of environmental errors (I/O, renaming file failures, etc): https://butler.cassandra.apache.org/#/ci/upstream/compare/Cassandra-4.0/cassandra-4.0 I started another another email thread a bit ago about keeping a releasable trunk; expect to see something on the list in the next week or two about concrete steps that came out of that discussion that we can take to try and mitigate this regression cycle. [Release progress] https://issues.apache.org/jira/secure/RapidBoard.jspa?rapidView=484&quickFilter=2278 4.0.2 Progress: 10 issues closed out on 4.0.2 in the past few weeks. Nothing too disruptive which is what we like to see; some tidying of config, error propagation, and flaky failure fixes. 4.1.0 Progress: 4.1.0 saw some additions in pursuit of CEP-10 and paxos support for the simulator. Other than that, some bits here and there; pretty quiet three weeks from a code perspective. This ebb in velocity isn't surprising going into the fall holiday season; we can probably expect to see things pick back up in early 2022. And that's a wrap for the past three weeks. Thanks everyone for the passionate and professional engagement on the project! ~Josh
