Health & Monitoring

Primentra exposes one health endpoint. Use it to confirm the service is up, the database is reachable, and the schema matches the application.

The endpoint

GET http://localhost:3001/api/health

The endpoint needs no authentication, so a monitoring system can poll it without an account. The installer uses it too: after registering the service it polls /api/health for up to 20 seconds and reports a failure if nothing answers.

Response fields

A healthy installation answers like this:

{
  "status": "ok",
  "database": "connected",
  "tablesExist": true,
  "schemaVersion": "1.2026.8.14",
  "ftsInstalled": true,
  "ftsServiceRunning": true,
  "ftsWindowsService": true,
  "fullTextSearch": true,
  "ftsIndexCount": 2,
  "appVersion": "1.2026.8.14"
}

Field	Meaning
`status`	`ok` when the database answered, `error` when it did not
`database`	`connected` or `disconnected`
`tablesExist`	`true` when all nine core tables are present: Models, Entities, Attributes, EntityRows, EntityValues, Roles, Permissions, AuditLog, Users
`schemaVersion`	The version recorded in the `SchemaVersion` table, or `null` when the schema was never deployed
`appVersion`	The version of the running application
`ftsInstalled`	The Full-Text Search feature is installed in SQL Server
`ftsIndexCount`	Number of full-text indexes found on `EntityRows` and `EntityValues`. Two is the expected value
`ftsServiceRunning`	A real `CONTAINS` query succeeded, so the filter daemon answers
`ftsWindowsService`	The `MSSQLFDLauncher` Windows service reports RUNNING
`fullTextSearch`	The summary flag. `true` only when `ftsServiceRunning` is true and `ftsIndexCount` is at least 2

When the database cannot be reached, the response is short:

{
  "status": "error",
  "database": "disconnected",
  "message": "Failed to connect to localhost:1433"
}

Status codes

Code	Meaning
`200`	The endpoint answered. Read the `status` field to learn whether the installation is healthy
`429`	Rate limited. Public endpoints allow 200 requests per 15 minutes per client
No response	The Windows service is down, the port is wrong, or a firewall blocks it

A failing health check still returns HTTP 200. Alert on the status field, not on the status code alone.

What to monitor

`status` is not `ok` — the service runs but cannot reach SQL Server. Check the instance, the credentials, and logs\errorlog.txt.
`schemaVersion` differs from `appVersion` — the application was upgraded but the schema was not. Sign in as an administrator and complete the upgrade. See Upgrading.
`schemaVersion` is `null` or `tablesExist` is false — the schema was never deployed. Run first-run setup.
`fullTextSearch` is false while you expect it — most often the filter daemon service is stopped. See Install SQL Server.

How often to poll

Poll once a minute at most. The 200-request budget per 15 minutes is shared by every unauthenticated endpoint, including sign-in. A monitor that polls every second exhausts it and locks real users out of the login screen.

Other checks

Check	Where
Licence and trial state	`GET /api/trial-status` on the server itself — see Licensing
Windows service state	`sc query Primentra`
Application errors	`logs\errorlog.txt`, or Settings › Logs — see Error Log
Full-Text Search detail	General Settings › Connection — see Database Connection

Upgrading — what to do when the schema version lags behind
Troubleshooting — service, connection and licence errors
Error Log — where the detail behind a failed check lands
Database Connection — connection settings and the Full-Text Search panel

Upgrading

Troubleshooting

Ready to get started?

Start managing your master data with Primentra today.

View Pricing