Search for a command to run...
I am an engineering student pursuing a degree in Artificial Intelligence and Data Science at Datta Meghe College of Engineering. I have strong technical skills in Full Stack Web Development, as well as programming in Python and Java. I currently manage Doubtly's blog and am exploring job opportunities as an SDE. I am passionate about learning new technologies and contributing to the tech community.
No comments yet. Be the first to comment.
1️⃣ Think in SYSTEMS, not code Before writing any code, always answer these 4 questions: What problem am I solving? What inputs go in? What outputs come out? What can fail? 👉 If you can’t explain this in 5 lines, don’t code yet. Mental model I...
If you’ve ever wanted to build and test your Flutter Web app directly in the cloud, without installing anything locally — GitHub Codespaces is your best friend. Codespaces gives you a full VS Code + Ubuntu dev environment inside your browser.With a f...
Adding a new column in ClickHouse is a common task when evolving your schema. Whether it’s a metric, metadata, or a JSON field, ClickHouse supports flexible and efficient schema changes — even at scale. ⚙️ Basic Syntax ALTER TABLE table_name ADD COL...
The Mechanics of Machine Vision: How MLLMs Interpret Visual Data To effectively guide Multimodal Large Language Models (MLLMs), one must first understand the intricate processes by which they interpret visual and textual data. These models are not mo...
Monitoring isn’t optional when you’re running ClickHouse in production.
As developers, we’ve all been there:
Queries suddenly slow down
Disk fills up overnight
Background merges eat up CPU
And worst of all — you only find out after users complain
That’s exactly why I built the ClickHouse Monitoring Stack — a single script that gets you ClickHouse + Prometheus + Grafana up and running in minutes.
In this blog, we’ll walk through why monitoring matters, what the stack includes, and how you can get it running on your own server.
ClickHouse is insanely fast, but without visibility, you’re just guessing when things break. Here’s what proper monitoring gives you:
Query Insights → Track slow queries & optimize them before users notice.
Resource Metrics → Watch CPU, memory, and disk I/O usage.
Cluster Health → Replication lag, node availability, and shard balance.
Alerts → Know when something’s wrong before production explodes.
In short: Monitoring turns chaos into control.
| Component | Port | Purpose |
| ClickHouse | 8123 | Database HTTP access |
| Prometheus | 9090 | Scrapes & stores metrics |
| Grafana | 3000 | Dashboards & alerts |
| CH Exporter | 9363 | Exposes ClickHouse metrics |
This script installs everything automatically — no manual configs required.
Clone the repo
git clone https://github.com/Ajinkgupta/clickhouse-monitoring-stack.git
cd clickhouse-monitoring-stack
Make the script executable
chmod +x setup.sh
Run the installer
sudo ./setup.sh
Enter your ClickHouse credentials when prompted.
That’s it. The script handles installation, Prometheus config, and Grafana setup automatically.
Grafana → http://<server_ip>:3000 (login: admin / admin)
Prometheus → http://<server_ip>:9090
ClickHouse HTTP → http://<server_ip>:8123
Metrics Exporter → http://<server_ip>:9363/metrics
Login to Grafana
Go to Dashboards → Import
Enter ID: 14192 (popular ClickHouse dashboard)
Select Prometheus as datasource → Done ✅
Now you’ll see:
Query latencies
Insert/Select counts
Disk usage growth
Background merges & mutations
Cluster replication lag
Perfect for devs debugging performance or scaling issues.
Set alerts in Grafana → “Disk > 80%” or “Query latency > 2s”
Watch storage growth → plan capacity before it’s too late
Monitor query mix → SELECT vs INSERT balance matters
Integrate alerts → Send them to Slack or PagerDuty
Author: Ajink Gupta
Contact: ajink@duck.com
This project is 100% open source. Fork it, improve it, and share feedback.
As developers, we spend too much time firefighting issues that proper monitoring could have prevented.
With this stack, you don’t have to waste hours setting up Prometheus and Grafana — just run one script and you’re done.
👉 Give it a try: ClickHouse Monitoring Stack on GitHub