Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix(config): respect api_server.workers #3049

Merged
merged 2 commits into from Sep 29, 2022
Merged

fix(config): respect api_server.workers #3049

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
1245984
fix: use `api_server.workers` from config correctly
judahrand Sep 29, 2022
3453280
chore: set null to `~`
aarnphm Sep 29, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
5 changes: 3 additions & 2 deletions bentoml/_internal/configuration/containers.py
View file
Open in desktop
@@ -1,10 +1,10 @@
from __future__ import annotations

import os
import math
import uuid
import typing as t
import logging
import multiprocessing
from copy import deepcopy
from typing import TYPE_CHECKING
from dataclasses import dataclass
Expand All @@ -24,6 +24,7 @@
from ..utils import split_with_quotes
from ..utils import validate_or_create_dir
from ..context import component_context
from ..resource import CpuResource
from ..resource import system_resources
from ...exceptions import BentoMLConfigException
from ..utils.unflatten import unflatten
Expand Down Expand Up @@ -449,7 +450,7 @@ def access_control_options(
return filtered_kwargs

api_server_workers = providers.Factory[int](
lambda workers: workers or (multiprocessing.cpu_count() // 2) + 1,
lambda workers: workers or math.ceil(CpuResource.from_system()),
api_server_config.workers,
)

Expand Down
2 changes: 1 addition & 1 deletion bentoml/_internal/configuration/default_configuration.yaml
View file
Open in desktop
@@ -1,5 +1,5 @@
api_server:
workers: 1
workers: ~ # When this is set to null the number of available CPU cores is used.
timeout: 60
backlog: 2048
metrics:
Expand Down
8 changes: 4 additions & 4 deletions bentoml/serve.py
View file
Open in desktop
Expand Up @@ -289,7 +289,7 @@ def serve_http_production(
port: int = Provide[BentoMLContainer.http.port],
host: str = Provide[BentoMLContainer.http.host],
backlog: int = Provide[BentoMLContainer.api_server_config.backlog],
api_workers: int | None = None,
api_workers: int = Provide[BentoMLContainer.api_server_workers],
ssl_certfile: str | None = Provide[BentoMLContainer.api_server_config.ssl.certfile],
ssl_keyfile: str | None = Provide[BentoMLContainer.api_server_config.ssl.keyfile],
ssl_keyfile_password: str
Expand Down Expand Up @@ -442,7 +442,7 @@ def serve_http_production(
),
],
working_dir=working_dir,
numprocesses=api_workers or math.ceil(CpuResource.from_system()),
numprocesses=api_workers,
)
)

Expand Down Expand Up @@ -650,7 +650,7 @@ def serve_grpc_production(
port: int = Provide[BentoMLContainer.grpc.port],
host: str = Provide[BentoMLContainer.grpc.host],
backlog: int = Provide[BentoMLContainer.api_server_config.backlog],
api_workers: int | None = None,
api_workers: int = Provide[BentoMLContainer.api_server_workers],
reflection: bool = Provide[BentoMLContainer.grpc.reflection.enabled],
max_concurrent_streams: int
| None = Provide[BentoMLContainer.grpc.max_concurrent_streams],
Expand Down Expand Up @@ -808,7 +808,7 @@ def serve_grpc_production(
args=args,
use_sockets=False,
working_dir=working_dir,
numprocesses=api_workers or math.ceil(CpuResource.from_system()),
numprocesses=api_workers,
)
)

Expand Down
4 changes: 2 additions & 2 deletions bentoml_cli/serve.py
View file
Open in desktop
Expand Up @@ -45,7 +45,7 @@ def add_serve_command(cli: click.Group) -> None:
@click.option(
"--api-workers",
type=click.INT,
default=None,
default=BentoMLContainer.api_server_workers.get(),
help="Specify the number of API server workers to start. Default to number of available CPU cores in production mode",
envvar="BENTOML_API_WORKERS",
show_default=True,
Expand Down Expand Up @@ -249,7 +249,7 @@ def serve( # type: ignore (unused warning)
@click.option(
"--api-workers",
type=click.INT,
default=None,
default=BentoMLContainer.api_server_workers.get(),
help="Specify the number of API server workers to start. Default to number of available CPU cores in production mode",
envvar="BENTOML_API_WORKERS",
show_default=True,
Expand Down
4 changes: 2 additions & 2 deletions bentoml_cli/start.py
View file
Open in desktop
Expand Up @@ -63,7 +63,7 @@ def add_start_command(cli: click.Group) -> None:
@click.option(
"--api-workers",
type=click.INT,
default=None,
default=BentoMLContainer.api_server_workers.get(),
help="Specify the number of API server workers to start. Default to number of available CPU cores in production mode",
envvar="BENTOML_API_WORKERS",
)
Expand Down Expand Up @@ -295,7 +295,7 @@ def start_runner_server( # type: ignore (unused warning)
@click.option(
"--api-workers",
type=click.INT,
default=None,
default=BentoMLContainer.api_server_workers.get(),
help="Specify the number of API server workers to start. Default to number of available CPU cores in production mode",
envvar="BENTOML_API_WORKERS",
)
Expand Down