Browse Source

Meeting notes from 20220615

Meeting notes from 20220615
Mike Chang 3 years ago
parent
commit
0fc8ecae07
1 changed files with 65 additions and 0 deletions
  1. 65 0
      meetings/notes/sig-meeting-20220615.md

+ 65 - 0
meetings/notes/sig-meeting-20220615.md

@@ -0,0 +1,65 @@
+## Meeting Details
+
+- **Date/Time:** June 15, 2022 @ 05:00 pm UTC / 10:00 am PDT
+- **Location:** [Discord SIG-Build Voice Room](https://discord.gg/wDNAQmatpq)
+- **Moderator:** Deepak Sharma
+- **Note Taker** Mike Chang
+
+The [SIG-Build Meetings](https://github.com/o3de/sig-build/tree/main/meetings) repo contains the history past calls, including a link to the agenda, recording, notes, and resources.
+
+## SIG Updates
+
+**What happened since the last meeting?**
+
+- Migrated nightly pipeline to O3DE.
+
+## Meeting Agenda
+
+- Discuss agenda from proposed topics
+  - Migrated nightly pipeline: https://github.com/o3de/sig-build/issues/34
+  - Intermittent Tests RFC needs a review from Build-SIG - https://github.com/o3de/sig-testing/issues/43
+  - Proposed RFC - GitHub PR Workflow - https://github.com/o3de/sig-build/issues/44
+  - Triage the new issues in the queue - https://github.com/o3de/o3de/issues?q=is%3Aissue+is%3Aopen+label%3Asig%2Fbuild+label%3Aneeds-triage+no%3Aassignee. 
+
+## Agenda notes
+
+### Migrated nightly pipeline: https://github.com/o3de/sig-build/issues/34
+dshmz: Status on the nightly builds
+shirang: A set of the nightly jobs have been migrated. Next steps are to add ios and mac jobs, but at a reduced build cadance
+brian: There's a GHI on the mac and ios blockers
+
+### Intermittent Tests RFC needs a review from Build-SIG - https://github.com/o3de/sig-testing/issues/43
+mike: The codeowners work is already being investigated by shirang
+shirang: Correct, I'm working looking at assigning failures by codeowners, and another modification to look at the asset to see if there's an owner
+mike: Is there a github issue for this work?
+shirang: No, but I will make one now
+shirang: There's also an ask from sig-testing to run a periodic build. The failure rate for rapid multipule runs will influence the overall average.
+dshmz: Work with sean to discuss the rate of failure is a concern
+
+### Proposed RFC - GitHub PR Workflow - https://github.com/o3de/sig-build/issues/44
+
+dweiss: Question about rate limiting, what methods do we have to handle failures? How do we handle situations where commands are failing to be processed?
+mike: The fallback is to go to the Jenkins job and run it manually
+shirang: For rate limiting, we could just check if there's a build already in progress and not process any further commands until it is done. We can also throw a comment that the command was rejected
+mike: That's a good point about rejection commands. We should have some sort of response to the commands, but maybe not successes
+brian: Yes, I was looking to reduce the amount of noise, so I was looking into just the failure response, but leave successes evident in the check
+mike: The one issue with rejecting commands on a current build is situations where a current build is having intermittent failures. Someone could be checking in code while a build is happening and wants to queue another one
+
+## Outcomes from Agenda
+
+Brian - Followup on the GHI on mac and ios builds for nightly with shirang
+Shirang - Comment on the RFC for intermittent tests on any existing work from sig-build and reach out to sean on periodic build failure concerns
+Brian - Update Github PR RFC following feedback
+
+#### SIG reviewers and maintainers groups
+- @brianherrera 
+- @amzn-changml 
+
+## Action Items
+
+- Chairs: No action
+- Maintainers/Reviewers: As noted in the agenda outcome
+
+## Open Discussion Items
+
+List any additional items below!