fca9167883
All checks were successful
Test and Publish Templates / test-and-publish (push) Successful in 9s
44 lines
1.5 KiB
Bash
Executable File
44 lines
1.5 KiB
Bash
Executable File
#!/bin/bash
|
|
source "${AGENT_PATH}/common.sh"
|
|
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
|
|
_check_required_env_vars "CONTAINER_NAME" "LLAMA_SERVER_URL" "MONITOR_PORT" "IMAGE_REGISTRY" "IMAGE_REPO" "IMAGE_TAG"
|
|
|
|
# Build GPU device mounts if nvidia-smi is available on the host
|
|
GPU_ARGS=""
|
|
if command -v nvidia-smi &>/dev/null; then
|
|
GPU_ARGS="--device /dev/nvidia0:/dev/nvidia0 \
|
|
--device /dev/nvidiactl:/dev/nvidiactl \
|
|
--device /dev/nvidia-uvm:/dev/nvidia-uvm \
|
|
-v /usr/bin/nvidia-smi:/usr/bin/nvidia-smi:ro \
|
|
-v /usr/lib/nvidia:/usr/lib/nvidia:ro \
|
|
-e LD_LIBRARY_PATH=/usr/lib/nvidia"
|
|
fi
|
|
|
|
DOCKER_RUN_CMD="docker run -d \
|
|
--restart unless-stopped \
|
|
--name ${CONTAINER_NAME} \
|
|
--network host \
|
|
${GPU_ARGS} \
|
|
-e LLAMA_SERVER_URL=${LLAMA_SERVER_URL} \
|
|
-e MONITOR_PORT=${MONITOR_PORT} \
|
|
-v ${SCRIPT_DIR}/monitor.py:/app/monitor.py:ro \
|
|
${IMAGE_REGISTRY}/${IMAGE_REPO}:${IMAGE_TAG} \
|
|
python3 /app/monitor.py"
|
|
|
|
echo "Starting container ${CONTAINER_NAME}..."
|
|
|
|
if ! _create_and_start_container "$DOCKER_RUN_CMD" "$CONTAINER_NAME"; then
|
|
if _is_container_exists $CONTAINER_NAME; then
|
|
echo "Attempting to get logs from failed container..."
|
|
_get_container_logs $CONTAINER_NAME
|
|
fi
|
|
_die "Failed to start container ${CONTAINER_NAME}"
|
|
fi
|
|
|
|
if ! _is_container_running "$CONTAINER_NAME"; then
|
|
_get_container_logs $CONTAINER_NAME
|
|
_die "Container ${CONTAINER_NAME} is not running after start attempt"
|
|
fi
|
|
|
|
echo "Service ${CONTAINER_NAME} started on port ${MONITOR_PORT}."
|