10 months ago
I’m experiencing an issue with my project where all services (Redis, PostgreSQL, and a FastAPI backend) stop running after a few hours, even though Serverless is disabled in the settings. (All at the same time)
Here are the key details:
The services shut down gracefully (I see SIGTERM and normal shutdown logs).
There’s no memory spike — total RAM usage stays below 500MB.
No crash logs or OOM errors are visible in the logs.
The project does not restart automatically after a request.
Uvicorn is running on 0.0.0.0:8000, and everything works fine until it silently stops.
Could you please help me identify why this is happening and how to prevent these shutdowns?
Log for fast api instance:
Application started successfully
INFO: Application startup complete.
INFO: Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)
INFO: 100.64.0.2:61000 - "OPTIONS /api/v1/auth/init HTTP/1.1" 200 OK
Processing auth/init request
Starting session for user dada9da7-20ae-42c8-8746-cdc110598e8e
Existing user found with device_id e3a71806-dd38-4005-8df1-5304c607ab25, updated last_seen_at
Selected 3 situations for user dada9da7-20ae-42c8-8746-cdc110598e8e
Generated new token pair for user dada9da7-20ae-42c8-8746-cdc110598e8e
Auth tokens generated for user dada9da7-20ae-42c8-8746-cdc110598e8e
Started session f57dd3a9-d3b8-487e-b9a2-7fc6c89c436a for user dada9da7-20ae-42c8-8746-cdc110598e8e
INFO: 100.64.0.2:61000 - "POST /api/v1/auth/init HTTP/1.1" 200 OK
INFO: 100.64.0.2:61000 - "OPTIONS /api/v1/session/start HTTP/1.1" 200 OK
Session started successfully
INFO: 100.64.0.2:61000 - "POST /api/v1/session/start HTTP/1.1" 201 Created
Starting session for user dada9da7-20ae-42c8-8746-cdc110598e8e
Stopping Container
INFO: Shutting down
INFO: Waiting for application shutdown.
Shutting down HumorAI Backend...
INFO: Application shutdown complete.
INFO: Finished server process [1]
Pinned Solution
9 months ago
Such problems usually occur when you use the free version of rialway, the server goes into sleep mode to save resources. Then it wakes up immediately after the call (checked via webhook). You can try pinging with a third-party service so that your process does not fall asleep, but is always active.
6 Replies
10 months ago
It has happened two times this week.
10 months ago
Hello,
Your 30 day Trial period has expired, please upgrade to a paid plan to continue using the platform.
Best,
Brody
brody
Hello,Your 30 day Trial period has expired, please upgrade to a paid plan to continue using the platform.Best,Brody
10 months ago
Thank you for your reply.
The issue is it wasn't showing as expired till today or yesterday.
And problems appeared before that.
9 months ago
Such problems usually occur when you use the free version of rialway, the server goes into sleep mode to save resources. Then it wakes up immediately after the call (checked via webhook). You can try pinging with a third-party service so that your process does not fall asleep, but is always active.
littleparty
Thank you for your reply. The issue is it wasn't showing as expired till today or yesterday. And problems appeared before that.
9 months ago
Is it the free plan? Your account says HOBBY now so is this resolved?
9 months ago
If it still happens share the new logs or other changes. As @euro2222222 mentioned you can keep the service up always but that could still be an issue with free
Status changed to Solved chandrika • 9 months ago