Prometheus Integration

Prometheus with Alertmanager is a widely used open-source monitoring and alerting stack. By integrating Alertmanager with Echobell, you can receive instant mobile notifications (including time-sensitive or call-style alerts) whenever your alert rules fire.

Prerequisites

An Echobell account with at least one channel created
A running Prometheus + Alertmanager setup
Access to edit your Alertmanager configuration (alertmanager.yml)

Setup Overview

Create an Echobell channel
Configure notification templates (and optional Link Template)
Choose an integration method:
- Webhook via Alertmanager webhook receiver (recommended)
- Email via Alertmanager email receiver (alternative)
Create or update alert rules
Test and refine

Step-by-Step Guide

1) Create an Echobell Channel

Open the Echobell app
Create a new channel (e.g., "Prometheus Alerts")
Pick a color for easy identification

2) Configure Notification Templates

Alertmanager’s webhook payload contains structured fields like status, commonLabels, commonAnnotations, externalURL and an alerts array. You can reference these fields directly in Echobell templates.

Suggested templates:

Title template:

{{commonLabels.alertname}} - {{status}}

Body template:

🔔 Alert: {{commonLabels.alertname}}
📊 Severity: {{commonLabels.severity}}
🖥️ Instance: {{alerts[0].labels["instance"]}}
⏰ Starts: {{alerts[0].startsAt}}
ℹ️ Summary: {{commonAnnotations.summary}}
🧾 Description: {{commonAnnotations.description}}

Optional Link Template (Advanced Settings):

{{alerts[0].generatorURL || externalURL}}

This shows a direct link to the firing graph or dashboard inside your alert record. If generatorURL isn’t present, it falls back to externalURL.

3) Integration Method A — Webhook (Recommended)

In your Echobell channel, copy the Webhook URL from Triggers
In Alertmanager, add a webhook receiver that posts to that URL

Minimal alertmanager.yml example:

route:
  receiver: echobell-webhook
 
receivers:
  - name: echobell-webhook
    webhook_configs:
      - url: "https://hook.echobell.one/t/REPLACE_WITH_YOUR_TOKEN"
        send_resolved: true
        http_config:
          follow_redirects: true

Notes:

send_resolved: true will also notify when alerts resolve (status becomes resolved)
Echobell reads the JSON body as-is, so the templates above will render correctly

3 alt) Integration Method B — Email (Alternative)

If you prefer or require email, use Alertmanager’s email receiver and send to the channel’s bound email address.

In Echobell, copy the channel’s email address from the Triggers section
Configure an email receiver in Alertmanager and use simple templates

Example snippet:

receivers:
  - name: echobell-email
    email_configs:
      - to: "PASTE_CHANNEL_EMAIL_ADDRESS"
        send_resolved: true
        subject: "{{ .CommonLabels.alertname }} - {{ .Status }}"
        html: |
          <p><strong>Summary:</strong> {{ .CommonAnnotations.summary }}</p>
          <p>{{ .CommonAnnotations.description }}</p>
          <p>
            <a href='{{ (index .Alerts 0).GeneratorURL }}' target='_blank' rel='noopener'>Open in Prometheus</a>
          </p>

In Echobell templates for email triggers, you can use variables like subject, text, and html that Echobell provides automatically.

4) Create an Example Alert Rule

Here’s a small example rule to test the flow:

groups:
  - name: example
    rules:
      - alert: HighCPU
        expr: avg(rate(process_cpu_seconds_total[5m])) > 0.8
        for: 2m
        labels:
          severity: critical
        annotations:
          summary: "High CPU usage"
          description: "CPU has been over 80% for 2 minutes"

Once this alert fires, Alertmanager will send the event to your configured receiver and Echobell will notify subscribers to the channel.

Testing the Integration

Temporarily lower thresholds or create a test alert that fires quickly
Confirm you receive an Echobell notification when it fires
If using the Link Template, open the record and tap the link
Resolve the alert and verify you receive the resolution notification (if enabled)

Choosing Notification Types

Time Sensitive: critical production outages, paging-level incidents
Calling: severe outages where you must wake or interrupt focus modes
Normal: routine alerts, warnings, or informational notices

Best Practices

Keep templates concise but informative (alert name, severity, instance, summary)
Use the Link Template to jump straight to the relevant graph or dashboard
Group and route alerts in Alertmanager to avoid noise and fatigue
Include severity labels and use Echobell notification types to match urgency
Protect your Echobell webhook URL like a secret; rotate if exposed

Troubleshooting

If alerts aren’t appearing in Echobell:

Verify the correct webhook URL or channel email address
Ensure the Echobell channel is active and has subscribers
Check Alertmanager logs for delivery errors
Confirm your route in alertmanager.yml actually points to the intended receiver
Test with a simple one-off alert rule
If using email, ensure your SMTP settings work and emails are being delivered

Privacy and Data Handling

Echobell is privacy-first:

Notification content stays on your devices
Only minimal configuration data is stored on servers
Webhook URLs and channel email addresses should be treated as secrets
Echobell does not retain your alert payloads on its servers

Prometheus & Alertmanager Integration