Microsoft SQL Server Dashboard

Microsoft SQL Server Dashboard — Pure T-SQL

60+ panels | 8 sections | Zero Prometheus | No exporters needed

Monitor your SQL Server instances using only native T-SQL DMVs. Connect Grafana's built-in Microsoft SQL Server data source directly to your instance — no additional tools required.

Data Source Setup

1. Add a Microsoft SQL Server data source in Grafana
2. Point it to your SQL Server instance (host:1433)
3. Use a login with VIEW SERVER STATE and VIEW ANY DEFINITION permissions (or SA for testing)
4. Import this dashboard — done!

A Docker Compose quickstart is available to test the dashboard in under 2 minutes.

Sections

🔮 Overview

🩺 Health Score (composite 0–100 gauge) · Server info (DB, version, edition, server name, uptime) · Batch Requests/s · Transactions/s · Page Splits/s · Full Scans/s · Lock Waits/s · Active sessions by login · Connections by application · Session status breakdown · Network I/O per client · Start time

⚡ Query Performance

Top 10 slowest queries · Top 10 CPU-heavy queries · Top 10 I/O-heavy queries (logical reads) · Top 10 most executed queries · Top 10 physical reads queries · Top 10 write-heavy queries · Compilations/s & Re-Compilations/s & Cache Hit % · Plan cache by type (donut) · Currently running requests (live table) · Cache Hit / Connections / Active Transactions gauges

🖥 Server Performance & Waits

Top 15 wait types (excludes benign waits) · Lock distribution by mode/status · Signal vs Resource waits (CPU vs I/O pressure) · Running / Sleeping / Blocked / Pending I/O counts · Active blocking chains · Open transactions with duration · TempDB allocation · I/O stall per DB file · Scheduler health per CPU

🧠 Memory & Buffer

Page Life Expectancy gauge · Memory Grants Pending gauge · Buffer Pool breakdown (clean/dirty) · Memory overview (Total/Target/Available/Used) · Top 10 memory clerks · Buffer pool per database (donut) · Index usage per table (seeks/scans/lookups/updates)

🔄 Index Health

Index fragmentation with REBUILD / REORGANIZE / OK recommendations · Unused indexes (zero reads, ongoing write cost)

💿 Database Space

Database sizes · Transaction log space · File distribution (donut) · Top 15 tables by size · Backup history · Data overview (tables count, total rows, missing indexes) · Missing index suggestions from the optimizer

🔒 Security & Errors

Logins / Logouts / Deadlocks / Errors counters · Sysadmin role members (security audit) · Database states with recovery model and compatibility level

⏰ Jobs Monitoring

Job frequency (last 7 days) · Running jobs with duration · Scheduled jobs · Job history · Failed jobs with error messages

How It Works

All panels query SQL Server DMVs (Dynamic Management Views) directly via the Grafana MSSQL data source. No Prometheus, no exporters, no agents.

Key DMVs used: sys.dm_exec_query_stats · sys.dm_exec_sessions · sys.dm_os_performance_counters · sys.dm_os_wait_stats · sys.dm_os_buffer_descriptors · sys.dm_os_sys_memory · sys.dm_os_memory_clerks · sys.dm_os_schedulers · sys.dm_tran_locks · sys.dm_tran_active_transactions · sys.dm_io_virtual_file_stats · sys.dm_db_index_usage_stats · sys.dm_db_index_physical_stats · sys.dm_db_missing_index_details · sys.dm_exec_query_memory_grants · sys.dm_exec_connections · msdb.dbo.sysjobs · msdb.dbo.sysjobhistory

Full metrics reference: docs/metrics.md

Contributions welcome on GitHub — give it a ⭐️ if you find it useful!