BestPromptFinder

Add Useful Logging and Monitoring to Your Service

Quality, usefulness and confidence are AI evaluations out of 100 — not user ratings. Community results come from visitors' "worked / didn't work" votes below. How scoring works →

The prompt

Help me make my service observable without drowning in logs. Service: [what it does, language and framework] Hosting: [VPS / Docker / Kubernetes / serverless] Current logging: [print statements / basic logger / nothing] Tools I have or can use: [e.g. none yet / Grafana and Prometheus / a hosted log service] What goes wrong today: [e.g. slow requests we can't explain, silent job failures] Key code: [paste the main request handler or job runner] Give me: 1. A logging plan: structured JSON logs, which events to log at which level, and the fields every line should carry (timestamp, request ID, user or tenant ID where allowed, route, duration, outcome). 2. Code changes for my framework to add a request ID, timing and error logging in one place, not in every function. 3. Four to six metrics worth tracking (e.g. request rate, error rate, p95 latency, queue depth, job failures) and how to emit them. 4. Three alerts with sensible starting thresholds and what someone should do when each fires. Rules: never log passwords, tokens, full card or ID numbers, or message bodies with personal data; show how to redact them. Keep the code changes small and reviewable.

Why this prompt

• Hand-picked, editorially reviewed prompt
• Written for Claude, ChatGPT, Gemini

Source

Created by BestPromptFinder — editorial prompt, not user-generated. See our content & source policy.

Related Coding prompts