From dd6a8b7a7cfd19ba4f0371bebff7d992b8b8f269 Mon Sep 17 00:00:00 2001 From: Josh Bailey Date: Mon, 18 Dec 2023 07:38:27 +0000 Subject: [PATCH] healthcheck. --- docker/Dockerfile.cuda-torchserve | 2 ++ docker/Dockerfile.orin-torchserve | 2 ++ docker/Dockerfile.rocm-torchserve | 2 ++ docker/Dockerfile.torchserve | 2 ++ torchserve/healthcheck.sh | 7 +++++++ torchserve/install-common.sh | 1 + 6 files changed, 16 insertions(+) create mode 100755 torchserve/healthcheck.sh diff --git a/docker/Dockerfile.cuda-torchserve b/docker/Dockerfile.cuda-torchserve index 2e6d60d..f9fe7df 100644 --- a/docker/Dockerfile.cuda-torchserve +++ b/docker/Dockerfile.cuda-torchserve @@ -8,6 +8,8 @@ RUN --mount=type=cache,target=/root/.cache /torchserve/install-torchserve.sh --c RUN /usr/local/bin/torchserve --help COPY torchserve/config.properties /torchserve/config.properties COPY torchserve/torchserve-entrypoint.sh /torchserve/torchserve-entrypoint.sh +COPY torchserve/healthcheck.sh /torchserve/healthcheck.sh +HEALTHCHECK CMD /torchserve/healthcheck.sh || exit 1 ENTRYPOINT ["/torchserve/torchserve-entrypoint.sh"] # see Dockerfile.torchserve for example, but use diff --git a/docker/Dockerfile.orin-torchserve b/docker/Dockerfile.orin-torchserve index 8242383..b269b52 100644 --- a/docker/Dockerfile.orin-torchserve +++ b/docker/Dockerfile.orin-torchserve @@ -8,6 +8,8 @@ RUN --mount=type=cache,target=/root/.cache /torchserve/install-orin-torchserve.s RUN /usr/local/bin/torchserve --help COPY torchserve/config.properties /torchserve/config.properties COPY torchserve/torchserve-entrypoint.sh /torchserve/torchserve-entrypoint.sh +COPY torchserve/healthcheck.sh /torchserve/healthcheck.sh +HEALTHCHECK CMD /torchserve/healthcheck.sh || exit 1 ENTRYPOINT ["/torchserve/torchserve-entrypoint.sh"] # 0. Always start from a clean sdkmanager, installation of Ubuntu 20.04. DO NOT attempt an in-place JetPack upgrade, DO NOT try to install docker-ce, and DO NOT try to re-install/replace CUDA without re-imaging the entire system. diff --git a/docker/Dockerfile.rocm-torchserve b/docker/Dockerfile.rocm-torchserve index 265d113..8fd66e5 100644 --- a/docker/Dockerfile.rocm-torchserve +++ b/docker/Dockerfile.rocm-torchserve @@ -8,6 +8,8 @@ RUN --mount=type=cache,target=/root/.cache /torchserve/install-rocm-torchserve.s RUN torchserve --help COPY torchserve/config.properties /torchserve/config.properties COPY torchserve/torchserve-rocm-entrypoint.sh /torchserve/torchserve-entrypoint.sh +COPY torchserve/healthcheck.sh /torchserve/healthcheck.sh +HEALTHCHECK CMD /torchserve/healthcheck.sh || exit 1 ENTRYPOINT ["/torchserve/torchserve-entrypoint.sh"] # TODO: must build locally - 20+GB required, which is too much for Github. diff --git a/docker/Dockerfile.torchserve b/docker/Dockerfile.torchserve index 0c15780..51bf2c6 100644 --- a/docker/Dockerfile.torchserve +++ b/docker/Dockerfile.torchserve @@ -8,4 +8,6 @@ RUN --mount=type=cache,target=/root/.cache /torchserve/install-torchserve.sh RUN /usr/local/bin/torchserve --help COPY torchserve/config.properties /torchserve/config.properties COPY torchserve/torchserve-entrypoint.sh /torchserve/torchserve-entrypoint.sh +COPY torchserve/healthcheck.sh /torchserve/healthcheck.sh +HEALTHCHECK CMD /torchserve/healthcheck.sh || exit 1 ENTRYPOINT ["/torchserve/torchserve-entrypoint.sh"] diff --git a/torchserve/healthcheck.sh b/torchserve/healthcheck.sh new file mode 100755 index 0000000..23980b3 --- /dev/null +++ b/torchserve/healthcheck.sh @@ -0,0 +1,7 @@ +#!/bin/bash + +RESPONSE=$(curl localhost:8080/ping | jq .status) +if [ "${RESPONSE}" == '"Healthy"' ]; then + exit 0 +fi +exit 1 diff --git a/torchserve/install-common.sh b/torchserve/install-common.sh index a2b0048..4b28e88 100755 --- a/torchserve/install-common.sh +++ b/torchserve/install-common.sh @@ -1,2 +1,3 @@ #!/bin/sh +apt-get update && apt-get install wget jq pip install opencv-python pandas py-cpuinfo scipy seaborn thop