Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request updates the default gateway name to otterscale-llm-d-infra-inference-gateway and introduces a hardcoded fsGroup of 1000 for the model artifact job's security context. While the implementation is functional, it is recommended to make the fsGroup configurable rather than hardcoded to improve the operator's flexibility and compatibility across different Kubernetes environments and security policies.
| // so that kit push authenticates transparently without an explicit login step | ||
| func BuildJob(artifact *modelv1alpha1.ModelArtifact, kitImage string, labels map[string]string) *batchv1.Job { | ||
| env := buildEnvVars(artifact) | ||
| fsGroup := int64(1000) |
There was a problem hiding this comment.
Hardcoding the fsGroup to 1000 limits the operator's flexibility in environments with strict security policies (e.g., OpenShift) or when using custom kit images that expect a different GID. Since the kit-image is already configurable via a flag, consider making the fsGroup configurable as well to improve the operator's portability and compatibility with various cluster configurations.
2 participants
No description provided.