Setting Up ClickHouse Monitoring with clickhouse-exporter, Prometheus, and Grafana

This article explains how to set up a monitoring system for a ClickHouse single node or cluster using clickhouse-exporter, Prometheus, and Grafana.

Deploy exporter

First, obtain the source code and compile it on a macOS machine (or any environment with internet access) because the production Linux host cannot access external networks.

mkdir -p $GO_PATH/src/github.com/Percona-Lab

cd $GO_PATH/src/github.com/Percona-Lab

git clone https://github.com/Percona-Lab/clickhouse_exporter

cd $GO_PATH/src/github.com/Percona-Lab

GO111MODULE=off CGO_ENABLED=0 GOOS=linux GOARCH=amd64 go build clickhouse_exporter.go

./clickhouse_exporter -h

(shows usage)

Run the exporter with the ClickHouse HTTP endpoint:

./clickhouse_exporter -scrape_uri=http://clickhouse_server_ip:8123/ -log.level=info

Configure Prometheus by adding a job to prometheus.yml :

- job_name: 'clickhouse_exporter'
  scrape_interval: 10s
  static_configs:
    - targets:
      - "clickhouse_exporter_ip:9116"

Reload Prometheus configuration:

curl 'http://localhost:9090/-/reload' -X POST

Import the ClickHouse dashboard (ID 882) into Grafana, select the Prometheus data source, and visualize the metrics.

The article concludes that this setup provides a basic monitoring solution for ClickHouse, and suggests extending it for multi‑node clusters or finer‑grained metrics.