Enable non-org autoscaler to read scale-config from any repo #5767
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎ 1 Skipped Deployment
|
facebook-github-bot
added
the
CLA Signed
jeanschmidt
reviewed
Oct 16, 2024
terraform-aws-github-runner/modules/runners/lambdas/runners/src/scale-runners/scale-up.ts
Outdated
Show resolved
Hide resolved
jeanschmidt
reviewed
Oct 16, 2024
terraform-aws-github-runner/modules/runners/lambdas/runners/src/scale-runners/scale-up.ts
Outdated
Show resolved
Hide resolved
|
Overall I do like the approach :) |
jeanschmidt
approved these changes
Oct 18, 2024
|
I like that this will make things cleaner. For LF Canary though, maybe I missed it but we'll still need a duplicate scale-config though right? Since scale-config for canary adds a |
zxiiro
approved these changes
Oct 18, 2024
Okay I just saw the related ci-infra PR and confirmed that yes we do still have a separate canary config. |
pytorchmergebot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Oct 18, 2024
Removing just the LF canary scale config for now to test the changes in pytorch/test-infra#5767 Those changes have been deployed to prod and appear to be working, but this will be the final proof that it is in fact reading the test-config version of scale-config and not the pytorch/pytorch copy. Note: This will break the Scale config validation workflow on test-infra, but it's worth it since this test will be very short lived and that workflow only runs when someone modifies scale config Pull Request resolved: #138361 Approved by: https://github.com/wdvr
ZainRizvi
added a commit
that referenced
this pull request
Oct 21, 2024
Part of the workflow to move the LF scale-config.yml files from pytorch/pytorch to test-infra (details in #5767) This updates the validation script to no longer expect to update the pytorch/pytorch version of these scale configs. It resulted in many other aspects of that file becoming simpler as well.
pytorchmergebot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Oct 22, 2024
Final step of moving scale config files to test-infra repo. Details in pytorch/test-infra#5767 Scale configs are now read from test-infra. This PR is just cleaning up stale files Pull Request resolved: #138511 Approved by: https://github.com/clee2000
SamGinzburg
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Oct 28, 2024
Final step of moving scale config files to test-infra repo. Details in pytorch/test-infra#5767 Scale configs are now read from test-infra. This PR is just cleaning up stale files Pull Request resolved: #138511 Approved by: https://github.com/clee2000
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Enable putting scale-config in test-infra, even for non-organization runners. Before, only scale-configs for organization-scoped runners could be put in arbitrary repos
Significant change:
scale_config_repono longer defaults to test-infra. You now have to specify a value if you're using org runners. Non-org runners will default to the job's repo. pytorch-gha-infra and ci-infra will need to be updated accordingly.Testing: Deployed these changes to LF canary, terminated all lf-c instances, and verified that creating a new pull request there would still result in instances getting provisioned