Cleanup deployment scripts and unused samples - #626
Merged
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
This PR cleans up deployment tooling by removing deprecated environment knobs, updating docs accordingly, bumping the default llm-d release to v0.4.0, and deleting unused vLLM-emulator deployment samples under deploy/.
Changes:
- Remove
BENCHMARK_MODEand inference-sim image override environment variables from scripts/workflows/docs. - Bump the default llm-d release used by
deploy/install.shtov0.4.0. - Delete the
deploy/examples/vllm-emulator/sample manifests and update Kind emulator docs to use GuideLLM load generation.
Reviewed changes
Copilot reviewed 16 out of 16 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| deploy/openshift/install.sh | Removes OpenShift-specific env overrides that are no longer used. |
| deploy/openshift/README.md | Drops BENCHMARK_MODE from documented environment variables. |
| deploy/kubernetes/README.md | Drops BENCHMARK_MODE from documented environment variables. |
| deploy/kind-emulator/install.sh | Removes unused inference-simulator image override env vars and benchmark mode. |
| deploy/kind-emulator/README.md | Removes vLLM emulator references and documents GuideLLM job-based load generation. |
| deploy/install.sh | Bumps llm-d release to v0.4.0 and removes benchmark-mode handling and sim image overrides; tweaks verification messaging. |
| deploy/examples/vllm-emulator/vllme-setup/vllme-variantautoscaling.yaml | Deletes unused vLLM emulator sample manifest. |
| deploy/examples/vllm-emulator/vllme-setup/vllme-deployment-with-service-and-servicemon.yaml | Deletes unused vLLM emulator sample manifest. |
| deploy/examples/vllm-emulator/prometheus-operator/prometheus-tls-values.yaml | Deletes unused vLLM emulator Prometheus TLS values. |
| deploy/examples/vllm-emulator/prometheus-operator/prometheus-deploy-all-in-one.yaml | Deletes unused vLLM emulator Prometheus operator manifest. |
| deploy/examples/vllm-emulator/integration_llm-d/vllme-inferencemodel.yaml | Deletes unused vLLM emulator llm-d integration manifest. |
| deploy/examples/vllm-emulator/integration_llm-d/arm64-gaie-sim-values.yaml | Deletes unused vLLM emulator ARM64 GAIE sim values. |
| deploy/examples/vllm-emulator/deploy.sh | Deletes unused vLLM emulator helper script. |
| deploy/examples/vllm-emulator/README.md | Deletes unused vLLM emulator documentation. |
| deploy/README.md | Removes BENCHMARK_MODE from the configuration reference. |
| .github/workflows/ci-e2e-openshift.yaml | Removes BENCHMARK_MODE usage from OpenShift E2E workflow env. |
Comment on lines
654
to
+658
| kubectl get variantautoscaling -n $LLMD_NS -o wide | ||
| fi | ||
| else | ||
| log_info "No VariantAutoscaling resources deployed yet (will be created by Helm chart)" | ||
| log_warning "No VariantAutoscaling resources found" | ||
| all_good=false |
There was a problem hiding this comment.
The VariantAutoscaling verification can silently pass when the CRD exists but there are zero VariantAutoscaling resources. kubectl get variantautoscaling exits 0 even when it prints "No resources found", so the else branch won’t run and all_good may remain true. Handle the va_count == 0 case explicitly (log warning and set all_good=false).
Collaborator
|
LGTM |
This file contains hidden or 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
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.
Per title, this PR:
llm-dversion tov0.4.0./deploydeployment directory.