If your team already runs Zabbix, you do not need to install the InfraCue Collector as well. Zabbix can forward the problems it already detects to Network Pulse as signed webhooks, and those events become device status, tickets and SLA timers in the same place as everything else.
InfraCue never connects into Zabbix and never stores a Zabbix credential. Traffic only ever goes outbound, from your Zabbix server to us.
This has been tested end to end against Zabbix 7.0. It needs Zabbix 6.0 or newer, because the signing uses the hmac() function built into the Zabbix JavaScript engine.
Step 1 — Create the connection in InfraCue
- Open the Integrations tabNetwork Pulse → Integrations.
- Name the integrationSomething you will recognise later, such as Zabbix — HQ.
- Choose Zabbix as the sourceThen select Create signed endpoint.
- Copy the credentials before you leave the pageThe endpoint, the integration key and the signing secret. The secret is shown once and is never displayed again.
Step 2 — Add the media type in Zabbix
- Create a media type in ZabbixGo to Alerts → Media types and select Create media type.
- Name it InfraCue and set Type to WebhookThe name is what you will pick later when you point an action at it.
- Paste the InfraCue script into the Script boxCopy it from infracue.js.
- Add the parameters listed belowThe first three are the values you copied from InfraCue; the rest are standard Zabbix macros.
- Tick Enabled, then saveA media type that is saved but not enabled delivers nothing.
| Parameter | Value |
|---|---|
endpoint |
the ingest URL from InfraCue, including ?company= and your workspace slug |
integration_key |
the integration key |
signing_secret |
the signing secret |
host |
{HOST.HOST} |
host_name |
{HOST.NAME} |
host_ip |
{HOST.IP} |
event_id |
{EVENT.ID} |
event_name |
{EVENT.NAME} |
trigger_severity |
{EVENT.SEVERITY} |
event_value |
{EVENT.VALUE} |
event_date |
{EVENT.DATE} |
event_time_only |
{EVENT.TIME} |
message |
{ALERT.MESSAGE} |
Step 3 — Send alerts to it
- Add the media to a userGo to Users → Users, open the user that should deliver these alerts, and add InfraCue under Media. Leave the severity range at the levels you want forwarded.
- Open or create a trigger actionAlerts → Actions → Trigger actions.
- Send to that user via InfraCueUnder Operations, add an operation that sends a message to the user through the InfraCue media type. Add the same under Recovery operations, or devices will go down in InfraCue and never come back up.
Trip a trigger you can control — or simply wait for the next real one — and the device should appear under Network Pulse → Devices.
What each Zabbix severity becomes
| In Zabbix | InfraCue severity | Device status |
|---|---|---|
| Disaster, High | critical | down |
| Average, Warning | warning | degraded |
| Information, Not classified | info | unchanged — see below |
| Any severity, problem resolved | info | up |
Information events do not change device status
Zabbix treats Information and Not classified as levels that are explicitly not outages — a configuration change, an agent restart, a login. InfraCue records those on the device's timeline but deliberately leaves its status alone.
That matters in both directions. If it did change status, an informational trigger firing while a Disaster was still open would downgrade the device from down to degraded; and that same trigger resolving would mark it up, taking a live outage off your wallboard.
If you want a particular informational trigger to drive device status, raise its severity in Zabbix to Warning or above. That setting is how you say "this is a health signal".
Before you rely on it
- Run the test command from the Integrations screen and confirm a device appears.
- Check that your Zabbix server clock is accurate. Signed requests are rejected if the timestamp is more than five minutes away from ours.
- Confirm recovery operations are configured, or devices will go down in InfraCue and never come back up.
- Set up alert rules under Network Pulse → Alert rules if you want these events to raise tickets. Until an admin configures the ticket defaults, alerting stays off and the screen says so.
If your notification template is customised
The adapter reads the field names Zabbix sends by default. Zabbix lets you author your own notification templates, so if yours has been changed, use the Generic signed webhook source instead and map your fields to the documented payload on your side. That path is described in forwarding any monitoring tool with a signed webhook.