mirror of
https://github.com/vale981/ray
synced 2025-03-10 05:16:49 -04:00
a9ede4e499
This change adds the GET, PUT, and DELETE commands for Serve’s REST API. The dashboard receives these commands and issues corresponding requests to the Serve controller.
435 lines
16 KiB
Python
435 lines
16 KiB
Python
from pydantic import ValidationError
|
|
import pytest
|
|
|
|
import requests
|
|
|
|
import ray
|
|
from ray.dashboard.modules.serve.schema import (
|
|
RayActorOptionsSchema,
|
|
DeploymentSchema,
|
|
ServeInstanceSchema,
|
|
deployment_to_schema,
|
|
schema_to_deployment,
|
|
serve_instance_to_schema,
|
|
schema_to_serve_instance,
|
|
)
|
|
from ray.util.accelerators.accelerators import NVIDIA_TESLA_V100, NVIDIA_TESLA_P4
|
|
from ray.serve.config import AutoscalingConfig
|
|
from ray import serve
|
|
|
|
|
|
class TestRayActorOptionsSchema:
|
|
def test_valid_ray_actor_options_schema(self):
|
|
# Ensure a valid RayActorOptionsSchema can be generated
|
|
|
|
ray_actor_options_schema = {
|
|
"runtime_env": {
|
|
"working_dir": (
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_module/archive/HEAD.zip"
|
|
)
|
|
},
|
|
"num_cpus": 0.2,
|
|
"num_gpus": 50,
|
|
"memory": 3,
|
|
"object_store_memory": 64,
|
|
"resources": {"custom_asic": 12},
|
|
"accelerator_type": NVIDIA_TESLA_V100,
|
|
}
|
|
|
|
RayActorOptionsSchema.parse_obj(ray_actor_options_schema)
|
|
|
|
def test_ge_zero_ray_actor_options_schema(self):
|
|
# Ensure ValidationError is raised when any fields that must be greater
|
|
# than zero is set to zero.
|
|
|
|
ge_zero_fields = ["num_cpus", "num_gpus", "memory", "object_store_memory"]
|
|
for field in ge_zero_fields:
|
|
with pytest.raises(ValidationError):
|
|
RayActorOptionsSchema.parse_obj({field: -1})
|
|
|
|
def test_runtime_env(self):
|
|
# Test different runtime_env configurations
|
|
|
|
ray_actor_options_schema = {
|
|
"runtime_env": None,
|
|
"num_cpus": 0.2,
|
|
"num_gpus": 50,
|
|
"memory": 3,
|
|
"object_store_memory": 64,
|
|
"resources": {"custom_asic": 12},
|
|
"accelerator_type": NVIDIA_TESLA_V100,
|
|
}
|
|
|
|
# ray_actor_options_schema should work as is
|
|
RayActorOptionsSchema.parse_obj(ray_actor_options_schema)
|
|
|
|
# working_dir and py_modules cannot contain local uris
|
|
ray_actor_options_schema["runtime_env"] = {
|
|
"working_dir": ".",
|
|
"py_modules": [
|
|
"/Desktop/my_project",
|
|
(
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_deploy_group/archive/HEAD.zip"
|
|
),
|
|
],
|
|
}
|
|
|
|
with pytest.raises(ValueError):
|
|
RayActorOptionsSchema.parse_obj(ray_actor_options_schema)
|
|
|
|
# remote uris should work
|
|
ray_actor_options_schema["runtime_env"] = {
|
|
"working_dir": (
|
|
"https://github.com/shrekris-anyscale/test_module/archive/HEAD.zip"
|
|
),
|
|
"py_modules": [
|
|
(
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_deploy_group/archive/HEAD.zip"
|
|
),
|
|
],
|
|
}
|
|
|
|
RayActorOptionsSchema.parse_obj(ray_actor_options_schema)
|
|
|
|
|
|
class TestDeploymentSchema:
|
|
def get_minimal_deployment_schema(self):
|
|
# Generate a DeploymentSchema with the fewest possible attributes set
|
|
|
|
return {
|
|
"name": "deep",
|
|
"init_args": None,
|
|
"init_kwargs": None,
|
|
"import_path": "my_module.MyClass",
|
|
"num_replicas": None,
|
|
"route_prefix": None,
|
|
"max_concurrent_queries": None,
|
|
"user_config": None,
|
|
"autoscaling_config": None,
|
|
"graceful_shutdown_wait_loop_s": None,
|
|
"graceful_shutdown_timeout_s": None,
|
|
"health_check_period_s": None,
|
|
"health_check_timeout_s": None,
|
|
"ray_actor_options": {
|
|
"runtime_env": None,
|
|
"num_cpus": None,
|
|
"num_gpus": None,
|
|
"memory": None,
|
|
"object_store_memory": None,
|
|
"resources": None,
|
|
"accelerator_type": None,
|
|
},
|
|
}
|
|
|
|
def test_valid_deployment_schema(self):
|
|
# Ensure a valid DeploymentSchema can be generated
|
|
|
|
deployment_schema = {
|
|
"name": "shallow",
|
|
"init_args": [4, "glue"],
|
|
"init_kwargs": {"fuel": "diesel"},
|
|
"import_path": "test_env.shallow_import.ShallowClass",
|
|
"num_replicas": 2,
|
|
"route_prefix": "/shallow",
|
|
"max_concurrent_queries": 32,
|
|
"user_config": {"threshold": 0.2, "pattern": "rainbow"},
|
|
"autoscaling_config": None,
|
|
"graceful_shutdown_wait_loop_s": 17,
|
|
"graceful_shutdown_timeout_s": 49,
|
|
"health_check_period_s": 11,
|
|
"health_check_timeout_s": 11,
|
|
"ray_actor_options": {
|
|
"runtime_env": {
|
|
"working_dir": (
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_module/archive/HEAD.zip"
|
|
),
|
|
"py_modules": [
|
|
(
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_deploy_group/archive/HEAD.zip"
|
|
),
|
|
],
|
|
},
|
|
"num_cpus": 3,
|
|
"num_gpus": 4.2,
|
|
"memory": 5,
|
|
"object_store_memory": 3,
|
|
"resources": {"custom_asic": 8},
|
|
"accelerator_type": NVIDIA_TESLA_P4,
|
|
},
|
|
}
|
|
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
def test_invalid_python_attributes(self):
|
|
# Test setting invalid attributes for Python to ensure a validation or
|
|
# value error is raised.
|
|
|
|
# Python requires an import path
|
|
deployment_schema = self.get_minimal_deployment_schema()
|
|
deployment_schema["init_args"] = [1, 2]
|
|
deployment_schema["init_kwargs"] = {"threshold": 0.5}
|
|
del deployment_schema["import_path"]
|
|
|
|
with pytest.raises(ValueError, match="must be specified"):
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
# DeploymentSchema should be generated once import_path is set
|
|
deployment_schema["import_path"] = "my_module.MyClass"
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
# Invalid import_path syntax should raise a ValidationError
|
|
invalid_paths = ["", "MyClass", ".", "hello,world"]
|
|
for path in invalid_paths:
|
|
deployment_schema["import_path"] = path
|
|
with pytest.raises(ValidationError):
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
def test_gt_zero_deployment_schema(self):
|
|
# Ensure ValidationError is raised when any fields that must be greater
|
|
# than zero is set to zero.
|
|
|
|
deployment_schema = self.get_minimal_deployment_schema()
|
|
|
|
gt_zero_fields = [
|
|
"num_replicas",
|
|
"max_concurrent_queries",
|
|
"health_check_period_s",
|
|
"health_check_timeout_s",
|
|
]
|
|
for field in gt_zero_fields:
|
|
deployment_schema[field] = 0
|
|
with pytest.raises(ValidationError):
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
deployment_schema[field] = None
|
|
|
|
def test_ge_zero_deployment_schema(self):
|
|
# Ensure ValidationError is raised when any fields that must be greater
|
|
# than or equal to zero is set to -1.
|
|
|
|
deployment_schema = self.get_minimal_deployment_schema()
|
|
|
|
ge_zero_fields = [
|
|
"graceful_shutdown_wait_loop_s",
|
|
"graceful_shutdown_timeout_s",
|
|
]
|
|
|
|
for field in ge_zero_fields:
|
|
deployment_schema[field] = -1
|
|
with pytest.raises(ValidationError):
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
deployment_schema[field] = None
|
|
|
|
def test_route_prefix(self):
|
|
# Ensure that route_prefix is validated
|
|
|
|
deployment_schema = self.get_minimal_deployment_schema()
|
|
|
|
# route_prefix must start with a "/"
|
|
deployment_schema["route_prefix"] = "hello/world"
|
|
with pytest.raises(ValueError):
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
# route_prefix must end with a "/"
|
|
deployment_schema["route_prefix"] = "/hello/world/"
|
|
with pytest.raises(ValueError):
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
# route_prefix cannot contain wildcards, meaning it can't have
|
|
# "{" or "}"
|
|
deployment_schema["route_prefix"] = "/hello/{adjective}/world/"
|
|
with pytest.raises(ValueError):
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
# Ensure a valid route_prefix works
|
|
deployment_schema["route_prefix"] = "/hello/wonderful/world"
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
# Ensure route_prefix of "/" works
|
|
deployment_schema["route_prefix"] = "/"
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
# Ensure route_prefix of None works
|
|
deployment_schema["route_prefix"] = None
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
def test_mutually_exclusive_num_replicas_and_autoscaling_config(self):
|
|
# num_replicas and autoscaling_config cannot be set at the same time
|
|
deployment_schema = self.get_minimal_deployment_schema()
|
|
|
|
deployment_schema["num_replicas"] = 5
|
|
deployment_schema["autoscaling_config"] = None
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
deployment_schema["num_replicas"] = None
|
|
deployment_schema["autoscaling_config"] = AutoscalingConfig().dict()
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
deployment_schema["num_replicas"] = 5
|
|
deployment_schema["autoscaling_config"] = AutoscalingConfig().dict()
|
|
with pytest.raises(ValueError):
|
|
DeploymentSchema.parse_obj(deployment_schema)
|
|
|
|
|
|
class TestServeInstanceSchema:
|
|
def test_valid_serve_instance_schema(self):
|
|
# Ensure a valid ServeInstanceSchema can be generated
|
|
|
|
serve_instance_schema = {
|
|
"deployments": [
|
|
{
|
|
"name": "shallow",
|
|
"init_args": [4, "glue"],
|
|
"init_kwargs": {"fuel": "diesel"},
|
|
"import_path": "test_env.shallow_import.ShallowClass",
|
|
"num_replicas": 2,
|
|
"route_prefix": "/shallow",
|
|
"max_concurrent_queries": 32,
|
|
"user_config": None,
|
|
"autoscaling_config": None,
|
|
"graceful_shutdown_wait_loop_s": 17,
|
|
"graceful_shutdown_timeout_s": 49,
|
|
"health_check_period_s": 11,
|
|
"health_check_timeout_s": 11,
|
|
"ray_actor_options": {
|
|
"runtime_env": {
|
|
"working_dir": (
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_module/archive/HEAD.zip"
|
|
),
|
|
"py_modules": [
|
|
(
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_deploy_group/archive/HEAD.zip"
|
|
),
|
|
],
|
|
},
|
|
"num_cpus": 3,
|
|
"num_gpus": 4.2,
|
|
"memory": 5,
|
|
"object_store_memory": 3,
|
|
"resources": {"custom_asic": 8},
|
|
"accelerator_type": NVIDIA_TESLA_P4,
|
|
},
|
|
},
|
|
{
|
|
"name": "deep",
|
|
"init_args": None,
|
|
"init_kwargs": None,
|
|
"import_path": ("test_env.subdir1.subdir2.deep_import.DeepClass"),
|
|
"num_replicas": None,
|
|
"route_prefix": None,
|
|
"max_concurrent_queries": None,
|
|
"user_config": None,
|
|
"autoscaling_config": None,
|
|
"graceful_shutdown_wait_loop_s": None,
|
|
"graceful_shutdown_timeout_s": None,
|
|
"health_check_period_s": None,
|
|
"health_check_timeout_s": None,
|
|
"ray_actor_options": {
|
|
"runtime_env": None,
|
|
"num_cpus": None,
|
|
"num_gpus": None,
|
|
"memory": None,
|
|
"object_store_memory": None,
|
|
"resources": None,
|
|
"accelerator_type": None,
|
|
},
|
|
},
|
|
]
|
|
}
|
|
|
|
ServeInstanceSchema.parse_obj(serve_instance_schema)
|
|
|
|
|
|
# This function is defined globally to be accessible via import path
|
|
def global_f():
|
|
return "Hello world!"
|
|
|
|
|
|
def test_deployment_to_schema_to_deployment():
|
|
@serve.deployment(
|
|
num_replicas=3,
|
|
route_prefix="/hello",
|
|
ray_actor_options={
|
|
"runtime_env": {
|
|
"working_dir": (
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_module/archive/HEAD.zip"
|
|
),
|
|
"py_modules": [
|
|
(
|
|
"https://github.com/shrekris-anyscale/"
|
|
"test_deploy_group/archive/HEAD.zip"
|
|
),
|
|
],
|
|
}
|
|
},
|
|
)
|
|
def f():
|
|
# The body of this function doesn't matter. It gets replaced by
|
|
# global_f() when the import path in f._func_or_class is overwritten.
|
|
# This function is used as a convenience to apply the @serve.deployment
|
|
# decorator without converting global_f() into a Deployment object.
|
|
pass
|
|
|
|
f._func_or_class = "ray.dashboard.modules.serve.tests.test_schema.global_f"
|
|
|
|
deployment = schema_to_deployment(deployment_to_schema(f))
|
|
|
|
assert deployment.num_replicas == 3
|
|
assert deployment.route_prefix == "/hello"
|
|
assert deployment.ray_actor_options["runtime_env"]["working_dir"] == (
|
|
"https://github.com/shrekris-anyscale/test_module/archive/HEAD.zip"
|
|
)
|
|
assert deployment.ray_actor_options["runtime_env"]["py_modules"] == [
|
|
"https://github.com/shrekris-anyscale/test_deploy_group/archive/HEAD.zip",
|
|
"https://github.com/shrekris-anyscale/test_module/archive/HEAD.zip",
|
|
]
|
|
|
|
serve.start()
|
|
deployment.deploy()
|
|
assert ray.get(deployment.get_handle().remote()) == "Hello world!"
|
|
assert requests.get("http://localhost:8000/hello").text == "Hello world!"
|
|
serve.shutdown()
|
|
|
|
|
|
def test_serve_instance_to_schema_to_serve_instance():
|
|
@serve.deployment(
|
|
num_replicas=1,
|
|
route_prefix="/hello",
|
|
)
|
|
def f1():
|
|
# The body of this function doesn't matter. See the comment in
|
|
# test_deployment_to_schema_to_deployment.
|
|
pass
|
|
|
|
@serve.deployment(
|
|
num_replicas=2,
|
|
route_prefix="/hi",
|
|
)
|
|
def f2():
|
|
pass
|
|
|
|
f1._func_or_class = "ray.dashboard.modules.serve.tests.test_schema.global_f"
|
|
f2._func_or_class = "ray.dashboard.modules.serve.tests.test_schema.global_f"
|
|
|
|
deployments = schema_to_serve_instance(serve_instance_to_schema([f1, f2]))
|
|
|
|
assert deployments[0].num_replicas == 1
|
|
assert deployments[0].route_prefix == "/hello"
|
|
assert deployments[1].num_replicas == 2
|
|
assert deployments[1].route_prefix == "/hi"
|
|
|
|
serve.start()
|
|
deployments[0].deploy()
|
|
deployments[1].deploy()
|
|
assert ray.get(deployments[0].get_handle().remote()) == "Hello world!"
|
|
assert requests.get("http://localhost:8000/hello").text == "Hello world!"
|
|
assert ray.get(deployments[1].get_handle().remote()) == "Hello world!"
|
|
assert requests.get("http://localhost:8000/hi").text == "Hello world!"
|
|
serve.shutdown()
|