See every ClickHouse query.
As it runs.
chmonitor turns your cluster's system tables into a fast, readable dashboard — queries, merges, parts, replication and health, plus an AI agent that answers questions about your data. Self-host it, or use the hosted dashboard.
Everything about your cluster, in one place
chmonitor reads ClickHouse system tables directly, no exporters, no extra storage. Each view is purpose-built so you can spot the problem and move on.
Ask your cluster anything
The built-in AI agent connects over MCP and answers questions about schemas, storage, query performance and health, then links you straight to the matching view.
- Chat about schema, storage, slow queries and system health
- Bring your own model, skills and MCP tools
- Read-only MCP endpoint for Claude, Cursor or any client
Every query, explained
Running, history, failed, slowest and most expensive queries, each with duration, memory, rows read and the full statement. Filter by user, slice by time and find the culprit fast.
- Queries by user and top query patterns over time
- Saved filter presets and one-click EXPLAIN
- Kill long-running queries right from the table
See the whole cluster at a glance
An interactive topology map of nodes, shards, replicas and the Keeper quorum, with live replication and coordination links, health states and an inspector for every cluster.
- ClickHouse nodes and Keeper quorum with leader & followers
- Physical and logical clusters, overlapping virtual clusters
- Per-node CPU, memory and latency, healthy / warn / unreachable
Explore tables and how they connect
Browse every database and table, map dependencies between them, then drop into a SQL editor to query the data, all without leaving the dashboard.
Catch issues early, then brief your agent to fix them
Real-time health indicators for the whole cluster, replication lag, failed queries, disk and memory, stuck mutations and more. When something turns red, generate a ready-to-paste prompt that briefs your AI agent to debug it.
Every health signal, color-coded
A single board of indicators, green, warn or critical at a glance, each linking straight to the views that explain it.
- Replication lag, readonly replicas and delayed inserts
- Failed and long-running queries, OOM kills, Keeper exceptions
- Disk, memory, parts-per-partition and mutation status
A ready audit prompt for your agent
One click turns any failing check into a structured prompt, the metric query, raw data, relevant system tables, common causes and docs. Paste it into Claude, ChatGPT or your coding agent for a tailored diagnosis and remediation plan.
- Toggle exactly what context to include
- Live values, thresholds, host and version baked in
- Copy as Markdown, works with any AI tool
One dashboard for the whole cluster
From a high-level overview down to ZooKeeper logs and replication mirrors, every system table you reach for, already laid out.
Overview & health
CPU, memory, connections and replication status at a glance, with health checks across the cluster.
Cluster insights
Record-breaking scans, busiest days, total rows read and storage, surfaced automatically from your query history.
PeerDB replication
Watch mirrors and peers with live status, rows-synced trend and routing, failures surface immediately.
Merges & parts
A live part-lifecycle timeline with merges, mutations, moves and part-size distribution by table.
Security & access
Users, roles, grants and quotas, review who can touch what, all from the same place.
Logs & system
Query, error, trace and ZooKeeper logs alongside settings and metrics, searchable and time-ranged.
Open source, built in public
Purpose-built for ClickHouse
General-purpose tools treat ClickHouse as one data source among many. chmonitor reads its system tables directly and tracks nothing else. Here's how it stacks up against the most common alternatives.
| chmonitor | Grafana + CH plugin | Datadog | CH native /dashboard | |
|---|---|---|---|---|
| Setup time | ~5 min (Docker or CF Workers) | Grafana + datasource plugin install | Agent on each ClickHouse node | Enable in ClickHouse config, restart |
| No exporters / agents needed | reads system tables directly | optional CH datasource plugin | agent required on each node | built into CH |
| ClickHouse-specific depth | Purpose-built: merges, parts, replication, query log, system health | Generic SQL panels, you write the queries | Generic agent metrics; limited CH specifics | Basic built-in panel; minimal detail |
| AI agent + MCP | built-in AI agent; MCP server at /api/mcp | none | AI Advisor on paid plans | none |
| Version-aware queries | SQL adapts to CH schema by version | you manage query compatibility | N/A | fixed to bundled version |
| Cost | Free self-host; cloud in early access | Grafana OSS free; Grafana Cloud paid | ~$15–23 per host per month (varies) | Free (bundled with ClickHouse) |
| Open source / self-host | GPL-3.0; Docker, CF Workers, K8s Helm | Grafana OSS (AGPL-3.0) | SaaS only | within ClickHouse |
Self-host free. Cloud for teams.
chmonitor is open source — self-hosting is always free under GPL-3.0. The hosted cloud connects a cluster without any setup. Cloud pricing is not finalised; early access is free to try and seats are limited.
- 1 ClickHouse host, 1 seat
- Full monitoring dashboard
- AI agent — 5 messages / day
- 7-day conversation & insights history
- Community support
- Everything in Free
- 3 hosts, 3 seats
- AI agent — 100 messages / day, then $5 / 2,000
- Scheduled AI Insights
- Basic alerting — up to 10 rules
- Anomaly detection + data export
- 30-day conversation & insights history
- Email support
- Everything in Pro
- 10 hosts, 10 seats
- AI agent — 1,000 messages / day, then $5 / 2,000
- Fleet view + advanced alerting (Slack, PagerDuty)
- Custom dashboards + webhook integrations
- API / MCP access
- 90-day conversation & insights history
- Priority support
- Everything in Max
- Unlimited hosts & seats
- Bring-your-own LLM key (BYOK)
- SSO / SAML, RBAC, audit logs
- Custom retention
- SLA & dedicated support
SELECT-only user for maximum safety.Common questions
Answers to what evaluators ask most before deploying chmonitor.
How do I deploy chmonitor?
docker run -d -p 3000:3000 -e CLICKHOUSE_HOST=... ghcr.io/chmonitor/chmonitor:latest. It also deploys to Cloudflare Workers and Kubernetes via the bundled Helm chart. See the quickstart for step-by-step instructions.Does it need exporters or agents installed on ClickHouse?
Is it safe to connect my production cluster?
SELECT queries against system tables and never writes to your data. The AI agent is also read-only. You control what the connection user can access.What is the performance impact on ClickHouse?
What license is chmonitor under?
What is the difference between self-host and cloud?
Which ClickHouse versions are supported?
More questions? Ask on GitHub Discussions.
Start monitoring in minutes
Open the hosted dashboard and connect a cluster, or grab the source and run it yourself. Self-hosting is always free.