How to Configure Readiness/Liveness Probes for Managed Omnigateway/Flexgateway on Runtime Fabric (BYOK)

Apply a gateway.mulesoft.com/v1alpha1 Configuration Kubernetes resource directly in the RTF app namespace where the Managed Omni Gateway is deployed. The resource must include the rtf.mulesoft.com/target-id label to target the correct gateway instance.

Step 1: Find the target-id

Inspect the gateway pod's environment variables:

FLEX_K8S_SELECTOR=rtf.mulesoft.com/target-id=<your-target-id>

Or check existing ApiInstance resources in the namespace — the rtf.mulesoft.com/target-id label will be present.

Step 2: Apply the Configuration YAML

apiVersion: gateway.mulesoft.com/v1alpha1
kind: Configuration
metadata:
  name: probe
  labels:
    rtf.mulesoft.com/target-id: <your-target-id>
  namespace: <rtf-app-namespace>
spec:
  probe:
    enabled: true
    port: 8888
    path: /healthz

Note: Ensure the path does not conflict with any existing API instance base paths. Also Ensure that Port number Which You are using is not used by existing Omni Gateway/Flex Gateway Service

Step 3: Verify

Test the readiness probe endpoint:

1. Login to Omni Gateway pod
   

   kubectl exec -it <Your Manged gateway Pod ID> -n <Your RTF App Namespace> -- /bin/sh

2. Hit the endpoint “http://localhost:8888/healthz/probes/readiness?allowEnvoyErrors=true&allowAPIErrors=true&allowPolicyErrors=true”
   Example:
   

   wget --spider -S http://localhost:8888/healthz/probes/readiness?allowEnvoyErrors=true&allowAPIErrors=true&allowPolicyErrors=true 2>&1 | grep "HTTP/"
   
   Output:
   
   HTTP/1.1 200 OK
     Content-Type: application/json
     Date: Sat, 30 May 2026 22:29:43 GMT
     Content-Length: 0
     server: Anypoint Flex Gateway
     connection: close

Note: RTF agent / Omni Gateway does not create a respective service or httproute/ingress object automatically. You can create service and httproute/ingress resources manually if required, based on your RTF / Loadbalancer / WAF setup requirements.

Notes & Limitations

Managed Omnigateway on CloudHub 2 (CH2): There is currently no supported way to configure probes. There is already a Feature request available in Ideas portal here [ https://ideas.salesforce.com/s/idea/a0BHp000017JiUkMAK/support-readinessliveness-probe-configuration-for-managed-omnigateway ], You can Upvote this Idea.

Workaround (Layer 7): Deploying a lightweight "dummy API" on the Flex Gateway can serve as a health check target for external load balancers — though this counts against API usage entitlements.

TCP Probe Alternative: For load balancers that support it, a TCP probe to port 443 can be used as a simpler alternative.

If you have any concerns or any queries on above setup, Please raise an Mulesoft Support case.