Been running Celery in production for years. Flower was always the default, and always the source of most of the pain. It falls over under load, the UI feels like 2015, and past a few hundred thousand tasks a day it stops being useful. So I built the thing I wished I had. Rust backend, Next.js dashboard. It chews through millions of events without eating memory, runs fine on a 2 GB box for small setups, and stays self-hosted because Celery payloads usually contain stuff (customer IDs, payment refs, API bodies) you really don't want sitting on someone else's dashboard. Things I'm happy with: - Real-time workflow DAGs (trace a chain of subtasks instead of guessing) - Autoscaled workers collapse into one row instead of 200 - Alert rules that don't page you at 3am for nothing - Heartbeat health that surfaces which pods are silently dying Apache 2.0, no telemetry, no calling home. If you've been stuck on Flower like I was, I'd love to hear what's missing. Issues and PRs welcome on GitHub.

Feloxi

Open-source Celery monitoring you self-host

Open-source Celery monitoring you self-host