mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-04-24 00:12:08 +00:00
Trying to avoid the random timeout. (#2929)
* Trying to avoid the random timeout. * More read timeout ? * Longer timeout ? * Remove legacy ENV directive. * Remove the dummy test, only increase the read timeout. * Wat?
This commit is contained in:
parent
17367438f3
commit
bdb3e488e4
@ -58,7 +58,7 @@ ARG INSTALL_CHANNEL=pytorch
|
|||||||
# Automatically set by buildx
|
# Automatically set by buildx
|
||||||
ARG TARGETPLATFORM
|
ARG TARGETPLATFORM
|
||||||
|
|
||||||
ENV PATH /opt/conda/bin:$PATH
|
ENV PATH=/opt/conda/bin:$PATH
|
||||||
|
|
||||||
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
|
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
|
||||||
build-essential \
|
build-essential \
|
||||||
|
@ -562,6 +562,7 @@ def launcher(event_loop):
|
|||||||
docker.types.DeviceRequest(count=gpu_count, capabilities=[["gpu"]])
|
docker.types.DeviceRequest(count=gpu_count, capabilities=[["gpu"]])
|
||||||
]
|
]
|
||||||
|
|
||||||
|
client.api.timeout = 1000
|
||||||
container = client.containers.run(
|
container = client.containers.run(
|
||||||
DOCKER_IMAGE,
|
DOCKER_IMAGE,
|
||||||
command=args,
|
command=args,
|
||||||
@ -573,7 +574,7 @@ def launcher(event_loop):
|
|||||||
devices=devices,
|
devices=devices,
|
||||||
volumes=volumes,
|
volumes=volumes,
|
||||||
ports={"80/tcp": port},
|
ports={"80/tcp": port},
|
||||||
healthcheck={"timeout": int(60 * 1e9), "retries": 2}, # 60s
|
healthcheck={"timeout": int(180 * 1e9), "retries": 2}, # 60s
|
||||||
shm_size="1G",
|
shm_size="1G",
|
||||||
)
|
)
|
||||||
|
|
||||||
|
Loading…
Reference in New Issue
Block a user