Add a workflow to update the viable/strict branch#7435
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/7435
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
osalpekar
left a comment
osalpekar
left a comment
There was a problem hiding this comment.
Looks great from my end! Will defer stamping to the vision maintainers.
pmeier
left a comment
pmeier
left a comment
There was a problem hiding this comment.
So the behavior this adds is to run a workflow every 30 minutes that checks whether all workflows are green and if so advances the viable/strict branch to the current main commit, right?
| uses: pytorch/test-infra/.github/workflows/update-viablestrict.yml@main | ||
| with: | ||
| repository: pytorch/vision | ||
| required_check: "build,cmake,lint,Build Linux,Build M1,Build Macos,Tests on Linux,Tests on macOS,Docs,Lint" |
There was a problem hiding this comment.
These seems to be the name fields of our other workflows, right? For example:
What happens if we rename one of them and forget to update here? Will this just basically disable the workflow or will we get notified somehow?
There was a problem hiding this comment.
The job will failed with an error message "missing required workflow". You can find all the logic in the test-infra repo.
There was a problem hiding this comment.
How can we surface this failure? IIUC, this will only visible under the "Actions" tab on GitHub, which we aren't using.
There was a problem hiding this comment.
That is a very good question! The main health indicator of the "Update viable/strict" workflow is "the time lag" - the difference between the last commit in the main and the viable/strict branch. You can find this information for PyTorch. The plan is to have a separate dashboard for domains with "time lag" and number of failed nightly builds (not done yet).
The automated process will look like this:
- Calculate the time lag between main and viable/strict.
- If the time lag is higher some threshold, send a notification to the team.
There was a problem hiding this comment.
If the time lag is higher some threshold, send a notification to the team.
Does "team" here mean releng or domain maintainers?
In any case, we need something like that setup before we actually use the viable/strict branch for the nightlies. I'm aware this does not happen in this PR, but just making sure this doesn't get lost in the future.
There was a problem hiding this comment.
Does "team" here mean releng or domain maintainers? - Yes
What is the plan here? Build the nightlies from |
|
This PR will not affect nightly builds. There is an internal cron job that triggers nightly builds. The plan is to switch nightly to |
pmeier
left a comment
pmeier
left a comment
There was a problem hiding this comment.
Thanks @DanilBaibak! Not for this PR, but before we use this system to run our nightly builds, we need to address #7435 (comment).
|
Hey @DanilBaibak! You merged this PR, but no labels were added. The list of valid labels is available at https://github.com/pytorch/vision/blob/main/.github/process_commit.py |
Reviewed By: vmoens Differential Revision: D44416611 fbshipit-source-id: dc48e48b25225ea8fa32a8ec0611d3f7bf16dd20
5 participants
Add a workflow to update the viable/strict branch. The current changes will not affect building nightly builds for now.