Overview
Runs the upstream node_exporter on every node, exposing the machine's CPU, memory, disk and network metrics for Prometheus. The monitor/prometheus block scrapes it on every node at nodeExporterPort, 9100 by default.
- Shape
- Daemonset
- Reached at
- Port on every node
- Node package
pkgs.prometheus-node-exporter
Fetched on demand: the first time a block of this type is placed on a node, its agent fetches pkgs.prometheus-node-exporter, built from the node's own nixpkgs so the NixOS binary cache serves it. That first placement needs network access (or a binary cache on the LAN) and waits for the download; later placements on that node start at once.
Example
apiVersion: expanse.io/v1 kind: Block metadata: name: node-exporter namespace: default spec: type: monitor/node-exporter strategy: kind: DAEMONSET resources: requests: cpu: 50m memory: 64Mi config: port: 9100
$ expanse ctl block apply -f node-exporter.yaml --dry-run # validate only $ expanse ctl block apply -f node-exporter.yaml
Configuration
Keys go under spec.config and are checked against the type's JSON Schema when the block is applied. A default is what the block uses when the key is unset.
| Key | Default | Meaning |
|---|---|---|
port | none: always set it | TCP port of the metrics endpoint. Unset, the exporter takes a random free port, which Prometheus cannot find; set 9100. |
webTelemetryPath | /metrics | URL path of the metrics endpoint. |
Using it
$ curl -s http://<node address>:9100/metrics | head
When a node fails
A daemonset has one replica per node, so there is nothing to move: a failed node's exporter is gone until the node returns.
Limits
- Collector flags cannot be set.
Source
- nix/blocks/monitor/node-exporter: the type's identity and JSON Schema.