chmonitor started as a hosted dashboard. We ran the app, sold seats, metered usage.
That does not match who actually uses this. Most readers are engineers at companies that already have DevOps. They already run ClickHouse. Docker or Helm is a normal Tuesday. They are not going to pay monthly for a UI they can put next to the cluster.
Data is also a problem. Query text, table names, topology: that is company data. People who write in ask for allowlists, an isolated network, a binary they can inspect, something security will sign off on. A SaaS dashboard has to pass all of that before anyone logs in. Running it yourself does not.
Why we dropped the SaaS pitch
You can deploy it. The cluster is already on your network. Security will not open ClickHouse to a random vendor. And system.query_log is not something you copy off-site without a fight.
So we stopped selling “we host the dashboard.” Today the work goes into the product: pages, advisor, agent. If you need an invoice, buy a license. The app is still GPL-3.0. No key. We trust you on host count.
The offer
OSS is free. Unlimited hosts. No nag screen.
Paid licenses are sized by how many hosts you monitor:
| License | Hosts | Yearly | Lifetime |
|---|---|---|---|
| Personal Self Hosted | unlimited | $0 | — |
| Team | 3 | $499 | $1,349 |
| Unlimited | unlimited | $999 | $2,999 |
Lifetime is one payment if you do not want renewals.
A host is one connection (one ClickHouse endpoint, or one Postgres source). Replicas in the same shard are not counted.
How you buy it
Pay on Polar. We ask for company name and website. There is no license key that unlocks features. Use the host count you paid for.
Listing on the customers page is optional and off by default.
Pricing → Buy → Polar. Docs: commercial license.
dash.chmonitor.dev
Still there if you do not want to run the app (demo, side project, or a team that is allowed to send metadata out).
If ClickHouse is behind a firewall, run the dashboard next to it: