High RAM usage on app platform using console (Django/Python)

Question

Does any else experience RAM usage spiking when using the app platform console? I rarely use the console, but I notice my RAM usage jumps really high and never comes down unless I redeploy/restart.

I’m usually using python/Django’s shell and doing model checks or really small querysets in Django (rarely do anything large). Not surprising to see a short jump in RAM usage, but it’s sustained indefinitely. Sometimes the site becomes unreachable.

Is there a way to address this without redeploying each time?

alexdo · Answer

Heya, @phosphide Have you noticed the same experience if you use the console via the API or CLI? If you’re using the browser console, do you simply close the tab, the browser or actually exit the Python console? https://docs.digitalocean.com/products/app-platform/how-to/console/ Regards

KFSys · Answer

Heya, Not just you — this is a pretty common gotcha with the App Platform console. The console isn’t a separate sandbox; it opens a shell inside a running instance of your component , so whatever you run there counts against that same instance’s memory budget. Firing up manage.py shell loads your entire Django app into a Python process, and CPython is famously reluctant to hand memory back to the OS once it’s grabbed it — even after you’re done with a queryset. So the RAM climbs and just… stays there for the life of that process. The reason it only drops on redeploy is that the shell process (and anything still holding references, like Django shell’s _ last-result variable) sticks around until the container is recreated. If you close the browser tab without cleanly exiting, that session can linger too. On a smaller instance, that leftover shell eats into the headroom your web workers need, which is why the site sometimes goes unreachable. A few things that help without redeploying every time: Exit cleanly with exit() / Ctrl-D and close the console tab, then double-check nothing lingered with ps aux | grep python. Ideally, don’t run interactive shells against your live web component at all. Running one-off checks as a Job component, or against a separate worker/tools component, keeps that RAM off your web instance. If you do need the shell handy, giving the instance a bit more headroom (size up, or trim gunicorn worker count) means a transient spike won’t push it into an OOM. Short version: it’s not a leak in your code, it’s the console sharing your web instance’s memory and Python not releasing it. Moving that work off the web component is the clean fix. Hope that helps!

Become a contributor for community

DigitalOcean Documentation

Resources for startups and AI-native businesses

The developer cloud

Start building today

Report this

High RAM usage on app platform using console (Django/Python)

Related questions

Digitalocean APP for Django not update static

DigitalOcean App Django not update static files

How to Deploy a Django + Next.js Application on DigitalOcean App Platform

How to Use Redis for Caching and Session Management in Laravel, Django, and Node.js on Ubuntu 24.04 (Self-Hosted and DigitalOcean Managed Re

Django App & PostgreSQL Database Connectivity Between Two Droplets

Django vs Laravel: Which Framework Should You Choose?

Advanced Django CI/CD Pipeline with GitHub Actions

Django JWT + Next.js Authentication Tutorial

Django REST Framework vs Django Ninja: A Comprehensive Comparison for API Development

How to fully reset a dev database on App Platform

Django vs Flask: Which Python Framework for API Development to Choose.

Django/Nginx/Gunicorn "unknown errors" in nginx logs

Report this

High RAM usage on app platform using console (Django/Python)

Related questions

Digitalocean APP for Django not update static

DigitalOcean App Django not update static files

How to Deploy a Django + Next.js Application on DigitalOcean App Platform

How to Use Redis for Caching and Session Management in Laravel, Django, and Node.js on Ubuntu 24.04 (Self-Hosted and DigitalOcean Managed Re

Django App & PostgreSQL Database Connectivity Between Two Droplets

Django vs Laravel: Which Framework Should You Choose?

Advanced Django CI/CD Pipeline with GitHub Actions

Django JWT + Next.js Authentication Tutorial

Django REST Framework vs Django Ninja: A Comprehensive Comparison for API Development

How to fully reset a dev database on App Platform

Django vs Flask: Which Python Framework for API Development to Choose.

Django/Nginx/Gunicorn "unknown errors" in nginx logs