Skip to content

fix(serverless): reject an invalid --gpus-per-worker before upload - #127

Merged
Ryank90 merged 2 commits into
rc/serverlessfrom
fix/serverless-gpus-per-worker
Sep 24, 2026
Merged

Ryank90 merged 2 commits into
rc/serverlessfrom
fix/serverless-gpus-per-worker

Conversation

@Ryank90

@Ryank90 Ryank90 commented Sep 22, 2026

Copy link
Copy Markdown
Contributor

Summary

  • Reject --gpus-per-worker values other than 1, 2, 4, or 8 on deploy and apps scale before any request is sent.
  • Deploy checks the flag before packing and uploading the source archive, so a bad count no longer spends an upload and then fails with 422.

Test plan

  • go test ./internal/cmd/serverless/ -run 'ValidateGPUsPerWorker|WorkerConfigPatchFromFlags'
  • runware serverless deploy ./app.py --id my-app --gpu-type h100 --gpus-per-worker 3 fails locally and does not upload
  • --gpus-per-worker 2 and 4 are still accepted

@coderabbitai

coderabbitai Bot commented Sep 22, 2026 •

Copy link
Copy Markdown

Important

Review skipped

Auto reviews are disabled on this repository. To trigger a review, include coderabbit-review in the PR description. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Advanced

Run ID: 5e0c4cdb-baeb-4bce-8897-c7032e5fe766

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • 🔍 Trigger review

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@Ryank90
Ryank90 changed the base branch from rc/serverless to fix/serverless-rollout-help September 22, 2026 15:58
@Ryank90
Ryank90 force-pushed the fix/serverless-gpus-per-worker branch from 62a340b to 504905c Compare September 22, 2026 15:58
Comment thread internal/cmd/serverless/deploy.go Outdated
Comment thread internal/cmd/serverless/apps_scale.go Outdated
Comment thread internal/cmd/serverless/deploy_test.go
@Ryank90
Ryank90 force-pushed the fix/serverless-gpus-per-worker branch from 504905c to cd80f61 Compare September 22, 2026 18:29
Base automatically changed from fix/serverless-rollout-help to rc/serverless September 23, 2026 16:49
Create and scale only accept 1, 2, 4, or 8. Deploy was sending any integer
and learning that from a 422 after the source archive had already uploaded.
One slice drives the help, the error, and the local check, and deploy rejects an invalid value before it builds the archive.
@Ryank90
Ryank90 force-pushed the fix/serverless-gpus-per-worker branch from cd80f61 to d50116a Compare September 24, 2026 14:39
@Ryank90
Ryank90 merged commit 3dfb250 into rc/serverless Sep 24, 2026
4 checks passed
@Ryank90
Ryank90 deleted the fix/serverless-gpus-per-worker branch September 24, 2026 14:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants