Prometheus is the de‑facto standard for metrics and observability in modern systems. It powers dashboards, alerting pipelines, SLO monitoring, and real‑time analytics across cloud platforms, on‑prem servers, and IoT deployments. With Nodlyn's new Prometheus connector, these capabilities become first‑class citizens inside your workflows — without sacrificing the local‑first execution model.
Why Prometheus Fits Perfectly in Local-First Automation
Prometheus is designed for reliability, simplicity, and independence. It runs on developer laptops, edge devices, Kubernetes clusters, and bare‑metal servers. This makes it a natural match for Nodlyn's philosophy: your workflows run locally, your metrics stay local, and your automation remains under your control.
Instead of routing metrics or alerts through a cloud automation platform, Nodlyn connects directly to your Prometheus instance — whether it's running on localhost, a Docker container, or a private monitoring server.
The Prometheus Connector: What It Can Do
The connector exposes three core capabilities that cover the most common observability workflows:
- scrapeMetrics — read raw metrics from any
/metricsendpoint - queryPromQL — execute PromQL instant queries via the Prometheus HTTP API
- alertTrigger — start workflows automatically when Alertmanager sends a webhook
Each capability is implemented as a standard Nodlyn node, with schema‑driven settings and predictable outputs. No custom UI code is required — everything flows through the existing inspector and palette.
Scraping Metrics: Simple and Direct
The scrapeMetrics capability reads any Prometheus exposition endpoint:
http://localhost:9100/metrics
http://my-service:8080/metrics
http://edge-device.local/metricsThe node returns the raw text exactly as Prometheus would receive it. This makes it ideal for:
- validating exporters
- debugging metric formats
- feeding metrics into downstream transformations
- building custom alerting logic inside workflows
PromQL Querying: Full Access to Prometheus Data
The queryPromQL capability executes instant queries using the official Prometheus API:
GET /api/v1/query?query=rate(http_requests_total[5m])This unlocks powerful use cases:
- threshold checks (CPU, memory, latency)
- rate calculations
- histogram quantiles
- service health evaluations
- custom SLO/SLA logic
The node returns the full JSON structure from Prometheus, including resultType and result, making it easy to integrate with any downstream workflow logic.
Alertmanager Webhook Trigger: Observability → Automation
Prometheus alerts become workflow triggers through the alertTrigger capability. Alertmanager sends a POST request to Nodlyn, and the workflow starts instantly.
This enables:
- auto‑remediation workflows
- incident response automation
- real‑time notifications
- dynamic scaling or throttling logic
- custom alert routing
The trigger supports both standard Alertmanager payloads and simplified JSON formats. Nodlyn extracts:
- alertName
- status (firing/resolved)
- labels
- annotations
- startsAt / endsAt
Connection Settings: Minimal and Explicit
Prometheus nodes use simple, explicit settings:
- Metrics URL — for scraping
- Prometheus Base URL — for PromQL queries
- Webhook Path — for alert triggers
- Timeout — optional, per node
No credentials are required for standard Prometheus setups, keeping configuration lightweight and secure.
Dry-Run Support: Safe by Default
Every Prometheus executor implements Nodlyn's dry-run mode. This means:
- No HTTP requests are sent
- No alerts are processed
- No external systems are touched
Dry-run lets you validate workflows, inspect outputs, and debug pipelines without interacting with Prometheus.
Performance Characteristics
Because Prometheus is local and Nodlyn executes locally, operations are extremely fast:
- Scrape — milliseconds
- PromQL — milliseconds
- Alert trigger — instant
There is no cloud round-trip, no external API gateway, and no vendor throttling. Your automation runs at the speed of your hardware.
Cloud-Compatible, Still Local-First
All Prometheus capabilities are marked CloudCompatible. If you choose to run workflows in Nodlyn Cloud, Prometheus nodes behave exactly the same — connecting to your Prometheus instance over a secure channel.
But the philosophy remains unchanged: you decide where your metrics live.
The Bottom Line
Prometheus brings powerful observability data into Nodlyn workflows. Combined with Nodlyn's local-first execution model, you get an automation stack that is fast, private, and deeply integrated with your monitoring systems.
Whether you're scraping metrics, running PromQL queries, or reacting to alerts, Prometheus fits seamlessly into the Nodlyn ecosystem.