chmonitorFeatures
Explore every chmonitor feature — query monitoring, storage, cluster health, metrics, and AI insights — with per-feature access control.
chmonitor is organized around the operational questions you answer while running a ClickHouse cluster. Each feature maps to one or more UI routes and a feature id you use to control access or disable it.
All features are public and enabled by default
You only configure what you want to restrict or turn off. The exceptions are the AI agent (agent) and control actions (actions), which default to authenticated access.
Feature pages
Overview
Cluster health at a glance — active queries, memory, disk, and CPU.
Query monitoring
Live, historical, failed, slow, and expensive query analysis.
Tables & storage
Table sizes, disk usage, parts, replication, and storage breakdown.
Data explorer
Browse column definitions, dependencies, and projections for any table.
Merges & operations
Active merges, mutations, and background operation throughput.
Clusters & replication
Replication lag, queue size, and replica health across hosts.
Metrics & profiler
Server CPU, memory, IO metrics and ClickHouse profiling data.
Insights
AI-surfaced anomalies, performance patterns, and observations.
Health & alerting
Active alerts from the health sweep cron with severity thresholds.
Security & audit
Who is accessing ClickHouse, with what grants, and how often.
Logs
Server log entries from system.text_log and related tables.
Dashboard
Build a custom view by combining any built-in charts on one canvas.
MCP server
ClickHouse monitoring tools for AI clients via the Model Context Protocol.
Feature index
Every feature answers a specific operational question. Use the feature id to gate or disable it.
| Feature | What it answers | Feature id | Page |
|---|---|---|---|
| Overview | Is my cluster healthy right now? | overview | Overview |
| Query Monitoring | What queries are running or slow? | queries | Query Monitoring |
| Tables & Storage | How big are my tables? Where is disk going? | tables | Tables & Storage |
| Data Explorer | What columns/dependencies does this table have? | tables | Data Explorer |
| Merges & Operations | Are merges/mutations keeping up? | operations | Merges & Operations |
| Clusters & Replication | Is replication lagging? Are replicas healthy? | cluster | Clusters & Replication |
| Metrics & Profiler | What are the server's CPU/memory/IO metrics? | metrics | Metrics & Profiler |
| Insights | What patterns or anomalies does the AI surface? | insights | Insights |
| Health & Alerting | Are there active alerts? | health | Health & Alerting |
| Security & Audit | Who is accessing ClickHouse and how? | security | Security & Audit |
| Logs | What did the server log? | logs | Logs |
| Dashboard | Custom pinned metrics view | dashboard | Dashboard |
| AI Agent | Natural language questions about my cluster | agent | AI Agent |
| MCP Server | ClickHouse monitoring tools for AI clients | mcp | MCP Server |
| PeerDB | PeerDB replication pipeline status | peerdb | PeerDB |
| Browser Connections | Personal ad-hoc connections stored in the browser | — | Browser Connections |
| Actions | Row-level actions across data pages (kill query, optimize, etc.) | actions | — |
| Settings | In-app settings and configuration | settings | Settings |
| Authentication | Who can sign in and what each visitor may access | — | Authentication |
Controlling features
Disable or restrict any feature with environment variables:
CHM_FEATURE_AGENT_ENABLED=false
# Require authentication for a feature
CHM_FEATURE_AGENT_ACCESS=authenticated
# Restrict multiple features at once
CHM_AUTH_REQUIRED_FEATURES=agent,mcp,settings,actions
# Disable multiple features at once
CHM_DISABLED_FEATURES=peerdb,browser-connectionsOr use a config file for complex setups:
CHM_CONFIG_FILE=/etc/clickhouse-monitor/config.toml[features.agent]
access = "authenticated"
[features.mcp]
access = "authenticated"
[features.peerdb]
enabled = falseSee Feature Permissions for the full reference.