Quickstart

From install to your first health check in under 2 minutes

Prerequisites

• Python 3.9+
• A running Redis instance
• Celery 5.2+ workers (they don’t need to be healthy — Kanari will tell you if they’re not)

---

1. Install

pip install kanari

---

2. Generate a config file

kanari init

This creates kanari.yaml in the current directory with sensible defaults. If REDIS_URL or CELERY_BROKER_URL are already set in your environment, those values are written into the file automatically. The command also pings Redis and reports whether it’s reachable:

✅ Created kanari.yaml
   Redis detected at redis://localhost:6379/0

   Next steps:
   kanari doctor        # verify connectivity first
   kanari audit         # one-shot health check

Open kanari.yaml and update redis_url / celery_broker_url if needed, then continue.

---

3. Verify your setup

kanari doctor

Checks that the required libraries are installed, Redis is reachable, and your Celery workers are responding. If anything is wrong, it tells you exactly how to fix it. Automatically loads kanari.yaml if it’s in the current directory.

Fix what doctor reports before running audit.

---

4. Run a health check

kanari audit

🔍 Kanari Audit
════════════════════════════════════════════════════════════

✅ System: HEALTHY

Infrastructure
  ✅ Redis: connected
  ✅ Celery: connected (3 workers)

Workers
  Status   Worker       Slots   Note
  ✅       api-worker   2/8     online
  ✅       email-wrk    1/4     online
  ⚠️       job-worker   4/4     at capacity

Queues
  Status   Queue     Pending   Latency
  ✅       celery    3         unknown
  ✅       default   0         0s
  ✅       emails    12        unknown

Metrics
  📊 Saturation: 43.8% (7/16 slots, headroom: 9 slots)
  ⏱️  Max Latency: unavailable — enable with: KanariStampPlugin.install(app)
  📋 Total Pending: 15 tasks

════════════════════════════════════════════════════════════
✅ All systems healthy
Audit completed in 1.2s

That’s it for local monitoring. No account, no external service beyond what you already have.

Tip: If kanari.yaml is in your current directory, all commands load it automatically — no --config flag needed. If the file is somewhere else, use kanari audit --config /path/to/kanari.yaml.

---

5. Connect to Kanari for alerts (optional)

To get Slack and email alerts when something breaks, connect the agent to your Kanari account:

kanari login

Enter your email, click the magic link, and your API key is saved automatically.

Then configure where to send alerts:

kanari alerts configure --slack-webhook https://hooks.slack.com/...
kanari alerts configure --email [e-mail]

Then run the continuous monitoring agent:

kanari agent

The agent runs in the background, collects metrics every 15 seconds, and sends you alerts the moment something goes wrong.

---

Next steps

kanari doctor

Diagnose connection issues, missing libraries, and config errors in one command.

Enable latency tracking

Add one line to your Celery app to measure real queue wait times.

Run a deep audit

Check your Redis and Celery configuration for common misconfigurations.

CI/CD integration

Use exit codes to fail your pipeline when Celery issues are detected.