Capacity forecast
Know which disks and which database files will fill up, and when, before the server stops.
A full disk stops a file server, a database or a mail system, often on a Monday morning. The Capacity screen follows how full each volume and each database file is, and estimates the date it will be full.
What you need
| Measurement | Agent module | Frequency |
|---|---|---|
| Machine volumes | AssetMonitor | every hour |
| Database files | DBMonitor, for the declared instances | every hour |
The agent must be at version 2026.10.11.3 or later (see Manage agents). Only fixed disks are measured, not network drives or USB sticks. For SQL Server, each file is measured; for MariaDB, the size of each schema.
Reading the screen
Open Asset Monitor → Capacity (#/am/capacite).
The Volumes tab ranks volumes from most to least urgent: Severity, Machine, Volume, Usage, Free, Growth per day, date when 90 % reached, date when the volume will be Full, and Trend (rising, stable, falling). Clicking a row opens the volume's curve over 30 to 800 days, with the forecast line dotted.
The Databases tab does the same for database files: size, space used, Maximum size and the volume that holds the file. A file with no maximum size shows "unlimited": only the volume that holds it can then fill up.
The Alerts tab lists capacity incidents over 7, 30 or 90 days.
How the date is estimated
- Each day, FirstSI keeps the highest space used during the day.
- It draws the line that best fits the last 30 days.
- It extends that line from the latest measurement up to 90% and 100%.
It takes at least 7 days of measurements; before that, the screen shows "not enough data". A change of more than 10% of capacity in one day (disk enlarged, big cleanup) is a break: only the days that follow it are used for the calculation. A volume that is not growing shows "no growth", with no date.
The calculation is redone on every measurement and once a day. An administrator can run it again with Recalculate.
Alerts
| Situation | Severity |
|---|---|
| Full within 30 days | medium |
| Full within 14 days | high |
| Full within 7 days, or already 95% full | critical |
The CAPA-VOLUME rule applies to volumes. CAPA-BASE applies to files that have a maximum size (same thresholds, relative to that size), and to the volume that holds an instance's files when it will be full within 14 days or is over 95%.
The incident closes on its own when the estimated date moves back beyond 45 days, when growth stops, or after 7 days without a measurement. Between 30 and 45 days, it stays as it is. Its severity can go up, not down.
Volumes that will be full within 14 days, or are already 95% full, also appear in the To handle list on the dashboard.
Frequently asked questions
"No volume measurement received yet". The AssetMonitor module is not active on the machine, or the agent is too old.
The saturation date jumped after a cleanup. The break is the cause: the forecast starts again from the days after the cleanup, and waits until it has 7 of them.
A database file is at 99% of its used space, with no alert. With no maximum size, the file grows as long as the volume allows. Keep an eye on the volume that holds it.
Source: · FirstSI Docs · updated 2026-10-11