[Serve] slashes in deployment name cause actor failure

[crypdick]

What happened + What you expected to happen

When creating a Serve deployment with a name containing a slash (such as TextGenerationModel.options(name="huawei-noah/TinyBERT_General_4L_312D")) leads to actor failures. The error occurs because the slash in the name is likely being used as a path separator in log files.

Error: FileNotFoundError: [Errno 2] No such file or directory: '/tmp/ray/session_2025-03-11_16-02-48_264712_45753/logs/serve/replica_default_huawei-noah/TinyBERT_General_4L_312D_eo3vqu7d.log'

Versions / Dependencies

ray, version 2.43.0

Reproduction script

from ray import serve

# The problematic model name containing a slash
model_with_slash = "huawei-noah/TinyBERT_General_4L_312D"
model_without_slash = "distilgpt2"  # This one works fine

@serve.deployment
class MyModel:
    def __init__(self, model_name):
        self.model_name = model_name

    def __call__(self, text):
        return f"Generated text using {self.model_name}: {text}"


def main():
    serve.start()
    
    print("Creating deployment with model name that doesn't contain a slash (works):")
    working_app = MyModel.options(name=model_without_slash).bind(model_without_slash)
    
    print("\nCreating deployment with model name that contains a slash (fails):")
    try:
        failing_app = MyModel.options(name=model_with_slash).bind(model_with_slash)
        serve.run(failing_app)
    except Exception as e:
        print(f"Error occurred: {type(e).__name__}")
        print(f"Error message: {str(e)}")



if __name__ == "__main__":
    main()

Issue Severity

Low: It annoys or frustrates me.