RHACS Setup
The ServiceNow endpoint must already exist. If it does not, finish ServiceNow Setup first — policy attachment against a 404 endpoint is a silent misconfiguration.
Prerequisites
- RHACS Central reachable and healthy
- Admin role in RHACS (
Adminor equivalent) - ServiceNow endpoint URL from ServiceNow Setup → step 3
- ServiceNow integration user credentials
1. Create The Generic Webhook Notifier
In the RHACS console:
Platform Configuration → Integrations → Notifier Integrations → Generic Webhook → New Integration
| Field | Value |
|---|---|
| Integration name | ServiceNow |
| Endpoint | https://<instance>.service-now.com/api/rhacs/alert |
| Skip TLS verify | ❌ (leave off outside lab environments) |
| Enable audit log | Optional |
| Username | ServiceNow integration user |
| Password | Integration user password (or API token if using OAuth flow) |
Click Test before saving. A successful test lands a ServiceNow Incident
titled ACS policy violation: Test (UNKNOWN_SEVERITY) (the test payload
carries no policy name or severity).
2. Attach The Notifier To Policies
Platform Configuration → Policy Management
For each policy that should produce a ServiceNow Incident:
- Open the policy
- Action → Edit Policy
- In Notifications, add
ServiceNow - Save
Attach deliberately. Every attached policy that fires opens an Incident — see Dedup + Storm Control.
3. Recommended Starter Set
Start with policies that produce low-volume, high-signal alerts:
| Policy | Why |
|---|---|
| Kubernetes Actions: Exec into Pod | Rare, high-context events, easy to triage |
| Kubernetes Actions: Impersonate Privileged User | Classic privilege escalation signal |
| Fixable Severity at least Important | CVE pressure with a known fix |
| Secret Mounted as Environment Variable | Config drift catch, low volume per cluster |
Avoid attaching high-volume policies (Ubuntu Package Manager in Image,
broad CVE findings) to the ServiceNow notifier until you have storm control
in place.
4. Validate End-to-End
From a lab cluster, trigger a known policy:
oc exec -n gibson garbage -c kserve-container -- /bin/bash
Confirm:
- RHACS shows the violation under Violations
- ServiceNow shows a new Incident within seconds
- Incident
Descriptioncontains cluster, namespace, deployment, pod, container, command, user, groups
If the Incident is missing fields, go to Handler Script Reference and verify the payload field names against Webhook Payload Reference.
Related
- Severity Mapping — tune ITIL priority before broad rollout
- Exec-into-Pod Use Case — full operator flow from trigger to assigned Incident