Historical Metric Storage

Track trends, investigate incidents, and plan capacity with stored metric history.

Data Retention

Metrics are stored in a SQLite database with WAL mode for fast writes without blocking reads. Old data is automatically pruned on a schedule, keeping your database lean while retaining the history you need.

  • SQLite with WAL mode for performance
  • Automatic pruning of old data
  • Configurable retention periods
  • Efficient storage — minimal disk footprint
Metric Storage Engine SQLite + WAL Size 24 MB (14 servers) Retention: ● Fast metrics 48 hours ● Hourly rollups 90 days

Interactive Charts

View historical metrics as real-time canvas-based charts. Select time ranges to zoom in on specific periods, compare metrics side by side, and explore trends visually.

  • Canvas-based real-time charting (ScoutGraph)
  • Time range selection (1h, 6h, 24h, 7d, 30d)
  • Smooth animations and hover tooltips
  • Multiple metrics on one chart
CPU % — Last 24h 1h 6h 24h 7d 100 75 50 25 0 00:00 06:00 12:00 18:00 Avg: 38% · Peak: 72% · Current: 41%

Capacity Planning

Historical data reveals trends that real-time monitoring can't. Spot disk growth over weeks, memory creep over months, or CPU utilization patterns that predict when you'll need to scale — before you hit limits.

  • Spot disk growth trends before you fill up
  • Track memory creep over weeks and months
  • Identify CPU utilization patterns
  • Plan upgrades with data, not guesswork
Disk /home — 30 day trend 85% ⚠ Projected to hit 85% in ~12 days Currently 72% · Growth: 0.5%/day

Incident Investigation

When something goes wrong, historical metrics let you correlate events with metric spikes. Pinpoint exactly when a problem started, what changed, and how long it lasted — essential for post-incident reviews.

  • Correlate past events with metric spikes
  • Pinpoint when problems started
  • Compare before and after any change
  • Essential for post-incident reviews
Incident: 2026-02-19 14:00-14:45 Peak: 94% Duration: 45 minutes Root cause: runaway cron job

Ready to Try Server Scout?

Start monitoring your servers and infrastructure in under 60 seconds. Free for 3 months.

Start Free Trial