Backup monitoring in PRTG

See every backup job as a PRTG sensor, with one set of channels per tier: status, hours since the last good run, size and restore points. Warning and error limits come pre-set from the schedule you gave Servatum, so a late or failed backup turns the sensor yellow or red without any threshold setup in PRTG.

Two ways to connect

Pull suits almost everyone: it needs no firewall change, and if PRTG can't reach Servatum, the failed request shows up in PRTG.

What the sensor shows

One sensor per Servatum job. Each tier of that job gets these channels:

An example response for a job with one tier:

{
  "prtg": {
    "result": [
      { "channel": "Offsite: status", "value": 1, "unit": "Count",
        "limitmode": 1, "limitminerror": 1 },
      { "channel": "Offsite: hours since last good run", "value": 7, "unit": "TimeHours",
        "limitmode": 1, "limitmaxwarning": 25, "limitmaxerror": 28 },
      { "channel": "Offsite: last backup size", "value": 48211993600, "unit": "BytesDisk" },
      { "channel": "Offsite: restore points", "value": 34, "unit": "Count",
        "limitmode": 1, "limitminwarning": 30 }
    ],
    "text": "nightly-full: all tiers ok"
  }
}

Two PRTG rules shape this design:

• 50 channels per sensor. PRTG supports at most 50 channels on these sensors. With four channels per tier, one sensor per job stays well inside that.
• Limits are read on the first scan. PRTG takes limits from the JSON when the sensor first sees a channel. If you change a job's schedule in Servatum later, adjust the limit in the channel settings or re-add the sensor.

Setting it up

1. In Servatum, create a read-only token for PRTG and copy the job's PRTG URL.
2. In PRTG, add an HTTP Data Advanced sensor to any device and paste the URL.
3. Under HTTP Headers, choose custom headers and add Authorization: Bearer <token>.
4. Set the scanning interval. Five to fifteen minutes is plenty for backups.

URL:     https://api.servatum.io/v1/prtg/nightly-full
Header:  Authorization: Bearer srv_ro_xxxxxxxx

The token can only read status. It can't change jobs or see anything else in your account.

Push instead of pull

If you'd rather not have PRTG poll, add an HTTP Push Data Advanced sensor. PRTG gives it a URL in the form https://<probe>:5051/<token>; paste that into Servatum and we'll send the same JSON on a fixed interval you choose, not only when something changes.

Two things to know. The probe must be reachable from the internet on that port. And PRTG records 0 for any channel left out of a push, so Servatum always sends every channel together. PRTG's own "No Incoming Data" setting decides what happens if nothing arrives for two scanning intervals. Choose Switch to down status so silence is an alert; that's why Servatum pushes on a schedule rather than only on changes.

What ships when

PRTG is a trademark of Paessler GmbH. Servatum is not affiliated with Paessler.