AI/ML

Automate Incident Alerts and Logging with n8n

Overview

n8n monitors server health via Prometheus, logs incidents in a database, and sends instant alerts to Slack when issues like high CPU usage are detected.

Prerequisites

n8n installed.
Prometheus instance with server metrics.
PostgreSQL database for logging.
Slack workspace with incoming webhook.

Workflow Setup

1. Prepare n8n Environment

Navigate to http://localhost:5678 and log in.
Click “Workflows,” then “+ New” and name it (e.g., “Incident Logger”).
Click “+” to add nodes.

2. Trigger Node: Schedule

Node: Schedule

Settings:

Interval: Every 5 minutes

Output: Triggers workflow.

3. Monitor Node: HTTP Request

Node: HTTP Request

Settings:

URL: http://<prometheus-host>:9090/api/v1/query?query=avg(rate(node_cpu_seconds_total[5m]))
Method: GET
Response Format: JSON

Output: CPU usage data (e.g., {{ $json.data.result[0].value[1] }}).

4. Log Node: PostgreSQL

Node: PostgreSQL

Settings:

Credentials: Database credentials
Query: INSERT INTO incidents (timestamp, cpu_usage) VALUES (NOW(), {{ $node["HTTP Request"].json.data.result[0].value[1] }})

Output: Incident logged.

5. Alert Node: Slack

Node: Slack

Settings:

Credentials: Slack webhook URL
Channel: #incident-alerts
Text: High CPU usage detected: {{ $node["HTTP Request"].json.data.result[0].value[1] }}%
Condition: Execute if {{ $node["HTTP Request"].json.data.result[0].value[1] > 80 }}

Output: Team notified.