On this page

SIGNL4

The Integrator sends each alert to SIGNL4 through the team’s inbound webhook. One monitor = one SIGNL4 alert: the monitor’s recovery closes it. You need the team secret of the SIGNL4 team that should be alerted (or the whole webhook URL).

SIGNL4 works like PagerDuty: same switches, same default destination on the Integrator, same automatic close.

How Mugnsoft uses SIGNL4
API SIGNL4 inbound webhook — https://connect.signl4.com/webhook/<team secret>
Destination The team secret (or an integration secret), or the whole webhook URL. Only an https URL on a signl4.com host is accepted, so a typo never sends alerts elsewhere.
Alert ID X-S4-ExternalID = <probe>:<type>:<monitor>, for example probe-paris:url:web-shop_probe-paris. Every alert of the monitor refers to the same SIGNL4 alert.
Non-OK status X-S4-Status: new — raises the alert
Back to OK X-S4-Status: resolved — closes the alert — with On change, and also with On fail alone
Title The alert headline: [CRITICAL] web-shop · HTTP 503 · probe-paris
Content The message, the monitor type as X-S4-Service (it selects the SIGNL4 category), monitor, status, target, probe and check time as extra fields, and a Locate in 3D view link

Step 1: get the team secret in SIGNL4

  1. Sign in to the SIGNL4 web portal with a manager account.
  2. Open Teams and select the team that should receive Mugnsoft alerts.
  3. Copy the team secret. The team’s webhook URL is https://connect.signl4.com/webhook/<team secret>.

To route Mugnsoft to a dedicated integration of the team instead, add a Webhook integration in Integrations and use its secret or URL.

Official guides: SIGNL4 inbound webhook, Integration with Webhook.

Anyone with the secret can alert the team: treat it as a secret. Mugnsoft stores it encrypted and never replicates it to a webserver-front.

Step 2: configure Mugnsoft

Default for every monitor of an Integrator — Integrator → General → Notifications → SIGNL4: switch On fail and/or On change on, and paste the secret in Default team secret.

Per monitor — monitor → Actions → SIGNL4 card: switch the trigger(s) on. Leave Team secret empty to use the Integrator’s default, or paste the secret of another team. Applications, Sentinel agents and container hosts have the same card.

Save both forms with the disk icon.

Recommended triggers:

Goal On fail On change
Alert on a lasting failure, close automatically on (rule: e.g. error, after 2, for 1) off
Follow every status change off on

With On fail alone, the alert is raised when the rule is met and closed when the monitor is back to OK. A failure that is also a status change sends one event, not two.

A probe with no Integrator, a Sentinel agent and the Webserver (for applications) send their SIGNL4 alerts themselves, with the same rules.

Test it

From Mugnsoft — in the Integrator’s Notifications block, click ⚡ on the SIGNL4 card. The Integrator raises a test alert ([TEST] Mugnsoft notification test) with the secret shown in the field, or its saved default, and closes it at once. This alerts the team on duty: warn them first.

By hand, from the Integrator’s host:


# Raise a test alert
curl -s -X POST https://connect.signl4.com/webhook/YOUR_TEAM_SECRET \
  -H "Content-Type: application/json" \
  -d '{"Title":"Mugnsoft test alert","Message":"test","X-S4-ExternalID":"mugnsoft-test","X-S4-Status":"new"}'

# Close it
curl -s -X POST https://connect.signl4.com/webhook/YOUR_TEAM_SECRET \
  -H "Content-Type: application/json" \
  -d '{"X-S4-ExternalID":"mugnsoft-test","X-S4-Status":"resolved"}'

Each call should answer with an eventId (HTTP 200 or 201).

Then run the end-to-end test.

CSV import and export

Monitor exports end with s4WH;s4OnF;s4OnSC, and so does the application export. A file exported before these columns existed still imports: an existing monitor or application keeps its SIGNL4 settings. See Import and export.

Troubleshooting

The Integrator logs the SIGNL4 error (sendMsgSignl4 - [monitor] - all 3 attempts failed: <error>). A failed alert goes to the retry buffer like any other channel.

Symptom Cause and action
not a SIGNL4 team secret The field holds something else than a secret (spaces, a truncated value…). Copy the team secret again.
not a SIGNL4 webhook URL A URL was pasted that is not https://...signl4.com/.... Paste the secret alone, or the webhook URL as SIGNL4 shows it.
SIGNL4 webhook returned HTTP 4xx Wrong or revoked secret. Copy it again from the team and run the ⚡ test.
The test succeeds but nobody is alerted Nobody of the team is on duty, or the category/service is muted in SIGNL4. Check the alert in the SIGNL4 portal.
Alerts are never closed The recovery was held by Debounce status-change alerts on the probe and the monitor failed again before it was stable.
An alert closed in SIGNL4 comes back The monitor is still failing: its next alert raises a new SIGNL4 alert.
Several alerts for one outage One alert per monitor and probe by design: a site checked from three probes raises three alerts.

More in Troubleshooting.

Translations

Custom script