My app has a tensorflow model to load and it takes around 20 secs to load it, but after deploying it in app engine it automatically shutdown after 7 minutes into starting the app and restart again. While this restarting the app wont work. How to avoid this? And what is this "Error while closing socket"
logs:
A 2021-03-23T06:10:28Z [2021-03-23 06:10:28 +0000] [8] [INFO] Application startup complete.
A 2021-03-23T06:17:26Z [2021-03-23 06:17:26 +0000] [1] [INFO] Handling signal: term
A 2021-03-23T06:17:26Z [2021-03-23 06:17:26 +0000] [8] [INFO] Shutting down
A 2021-03-23T06:17:26Z [2021-03-23 06:17:26 +0000] [8] [INFO] Error while closing socket [Errno 9] Bad file descriptor
A 2021-03-23T06:17:26Z [2021-03-23 06:17:26 +0000] [8] [INFO] Waiting for application shutdown.
A 2021-03-23T06:17:26Z [2021-03-23 06:17:26 +0000] [8] [INFO] Application shutdown complete.
A 2021-03-23T06:17:26Z [2021-03-23 06:17:26 +0000] [8] [INFO] Finished server process [8]
A 2021-03-23T06:17:26Z [2021-03-23 06:17:26 +0000] [8] [INFO] Worker exiting (pid: 8)
A 2021-03-23T06:17:27Z [2021-03-23 06:17:27 +0000] [1] [INFO] Shutting down: Master
A 2021-03-23T06:20:07Z [2021-03-23 06:20:07 +0000] [1] [INFO] Starting gunicorn 20.0.4
App.yaml:
runtime: python
env: flex
runtime_config:
python_version: 3
automatic_scaling:
min_num_instances: 1
max_num_instances: 1
cpu_utilization:
target_utilization: 1
readiness_check:
app_start_timeout_sec: 1800
liveness_check:
path: "/liveness_check"
check_interval_sec: 30
resources:
cpu: 1
memory_gb: 6
disk_size_gb: 15
entrypoint: gunicorn -k uvicorn.workers.UvicornWorker app.main:app --timeout 1000