[bug]: High memory/RAM usage with dockerized Plane API

[whallin]

Is there an existing issue for this?

• [X] I have searched the existing issues

Current behavior

The Docker version of the Plane API (makeplane/plane-backend:latest) uses a high amount of RAM on startup and counting.

The API/back-end container doesn't have any deployment resource limits right now, but with the way things are going, that could change quickly. When the container is first started, the memory usage is set right away to 400-600MB, even though the user hasn't done anything yet.

If you leave the service running and let users interact with the platform—slowly, but sometimes quickly, the RAM usage goes up, but there are no visible signs of garbage collection or anything like that. Is this a possible memory leak or a task that uses a lot of RAM in the API/back-end that might not be optimized?

Our current setup for Plane has no more than 3 registered users, 1 workspace with 4 projects, and only a few created issues. Redis, Postgres, and MinIO resource usage seems fine by their standards. The high amount of memory is only applicable to the Plane back-end Docker image.

I can say for sure that the front-end and worker containers don't have the same problem with using a lot of resources.

EDIT: Another theory here; could it be an upstream dependency causing the high memory usage?

Steps to reproduce

1. Create a Plane instance following the instructions of Self Hosting | Plane & using docker-compose-hub.yml
2. Start your Plane stack
3. Monitor resource usage during startup & while using your Plane instance

Browser

Google Chrome

Version

Self-hosted