simple_grpc_model_control.py

#!/usr/bin/env python
# Copyright 2020-2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions
# are met:
#  * Redistributions of source code must retain the above copyright
#    notice, this list of conditions and the following disclaimer.
#  * Redistributions in binary form must reproduce the above copyright
#    notice, this list of conditions and the following disclaimer in the
#    documentation and/or other materials provided with the distribution.
#  * Neither the name of NVIDIA CORPORATION nor the names of its
#    contributors may be used to endorse or promote products derived
#    from this software without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
# EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
# PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR
# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
# PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
# OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

import argparse
import sys

import tritonclient.grpc as grpcclient
from tritonclient.utils import InferenceServerException

if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "-v",
        "--verbose",
        action="store_true",
        required=False,
        default=False,
        help="Enable verbose output",
    )
    parser.add_argument(
        "-u",
        "--url",
        type=str,
        required=False,
        default="localhost:8001",
        help="Inference server URL. Default is localhost:8001.",
    )

    FLAGS = parser.parse_args()
    try:
        triton_client = grpcclient.InferenceServerClient(
            url=FLAGS.url, verbose=FLAGS.verbose
        )
    except Exception as e:
        print("context creation failed: " + str(e))
        sys.exit(1)

    model_name = "simple"

    # There are eight models in the repository directory
    if len(triton_client.get_model_repository_index().models) != 8:
        print("FAILED : Repository Index")
        sys.exit(1)

    triton_client.load_model(model_name)
    if not triton_client.is_model_ready(model_name):
        print("FAILED : Load Model")
        sys.exit(1)

    # Request to load the model with override config in original name
    # Send the config with wrong format
    try:
        config = '"parameters": {"config": {{"max_batch_size": "16"}}}'
        triton_client.load_model(model_name, config=config)
    except InferenceServerException as e:
        if "failed to load" not in e.message():
            sys.exit(1)
    else:
        print("Expect error occurs for invalid override config.")
        sys.exit(1)

    # Send the config with the correct format
    config = '{"max_batch_size":"16"}'
    triton_client.load_model(model_name, config=config)

    # Check that the model with original name is changed.
    # The value of max_batch_size should be changed from "8" to "16".
    updated_model_config = triton_client.get_model_config(model_name)
    max_batch_size = updated_model_config.config.max_batch_size
    if max_batch_size != 16:
        print("Expect max_batch_size = 16, got: {}".format(max_batch_size))
        sys.exit(1)

    triton_client.unload_model(model_name)
    if triton_client.is_model_ready(model_name):
        print("FAILED : Unload Model")
        sys.exit(1)

    # Single path component longer than NAME_MAX must be rejected without
    # terminating the server (invalid-arg response, not process abort).
    long_path = "file:" + ("A" * 256)
    try:
        triton_client.load_model(
            model_name,
            config="{}",
            files={long_path: b"a"},
        )
    except InferenceServerException as e:
        # TODO: [TRI-958] StatusCode.INVALID_ARGUMENT is more appropriate here
        if e.status() != "StatusCode.INTERNAL":
            print(
                "FAILED: expected status 'StatusCode.INTERNAL' for oversized file parameter, "
                "got status={!r}".format(e.status())
            )
            sys.exit(1)
        if "failed to poll from model repository" not in e.message():
            print(
                "FAILED: expected 'failed to poll from model repository' for oversized file parameter, "
                "got message={!r}".format(e.message())
            )
            sys.exit(1)
    else:
        print("FAILED: expected error for oversized file parameter")
        sys.exit(1)

    if not triton_client.is_server_live():
        print("FAILED: server not live after rejected load")
        sys.exit(1)

    # Trying to load wrong model name should emit exception
    try:
        triton_client.load_model("wrong_model_name")
    except InferenceServerException as e:
        if "failed to load" in e.message():
            print("PASS: model control")
            sys.exit(0)

    print("FAILED")
    sys.exit(1)