Hey all 
I’m self-hosting n8n in queue mode (main + webhook + worker) with Traefik, Redis, and Postgres via Docker Compose.
We recently restored from a snapshot and temporarily removed centralized logging (Promtail + Loki + Grafana) to stabilize things. Now we want to reintroduce logging to catch workflow failures and debug issues.
Our previous setup used:
Promtail to ship Docker logs
Loki to store logs
Grafana for dashboarding
Is this still the best practice in 2025? Or should we consider something simpler or more efficient just to catch failed executions?
Please provide any guidance or recommendations would really appreciate it!
Thanks!