Configuration for trace logging

When requested by IBM support, trace logging can be configured in Event Endpoint Management. Trace logs are useful for debugging the system, as they capture more details about what is happening in Event Endpoint Management.

You can configure trace logging for the Event Manager instance, Event Gateway instance or in the Event Endpoint Management UI.

Various trace levels are available, but the most common levels that are used for troubleshooting are trace or debug. The value to set for trace logging is provided by IBM Support.

Configuring trace logging on the Event Manager pods

Trace logging can be configured for an Event Manager instance by editing the EventEndpointManagement custom resource.

Configuring trace logging by using the custom resource traceSpec setting

A pod restart is not required when you change the trace setting in the custom resource.

To enable trace logging by using the custom resource setting, update the spec.manager field in your EventEndpointManagement custom resource as follows:

# excerpt from Event Endpoint Management CRD
spec:
  manager:
    traceSpec: "<VALUE>"

Configuring trace logging by using the TRACE_SPEC environment variable

Note: Enabling trace logging through the environment variable causes a restart of the pod.

In versions earlier than 11.5.1, to enable trace logging through the environment variable, update env in your EventEndpointManagement custom resource as follows:

# excerpt from Event Endpoint Management CRD
spec:
  manager:
    template:
      pod:
        spec:
          containers:
            - name: manager
              env:
                - name: TRACE_SPEC
                  value: "<VALUE>"

To enable trace logging when you are deploying with an overridden image, update the env field in your EventEndpointManagement custom resource as follows:

# excerpt from Event Endpoint Management CRD
template:
  pod:
    spec:
      containers:
        - env:
            - name: TRACE_SPEC
              value: "<VALUE>"
          image: >-
            <image-address>
          name: manager

Configuring trace logging on the Event Gateway pod

Operator-managed gateways

To enable trace logging on an operator-managed gateway, update the spec field in your EventGateway custom resource as follows:

spec:
  traceSpec: "<VALUE>"

Kubernetes Deployment gateways

Trace logging can be configured for a Kubernetes Deployment Event Gateway instance by adding trace.spec=<package>:<trace level> to your gateway ConfigMap.

apiVersion: v1
kind: ConfigMap
metadata:
  name: "<gateway group>-<gateway id>-config"
  labels:
    app: "testexample"
    gatewayGroup: "<gateway group>"
    gatewayId: "<gateway id>"
data:
  gateway.properties: |
    trace.spec=<value>

For more information about updating the gateway ConfigMap, refer to the gateway properties reference.

A pod restart is not required when you update trace.spec in the ConfigMap.

Docker gateways

On a Docker gateway, restart the gateway with a docker run command that includes the TRACE_SPEC argument. For example, docker run ... -e TRACE_SPEC=<VALUE>

In the Event Endpoint Management UI

It is possible to configure trace specifically for actions completed in the Event Endpoint Management UI. UI trace is configured by adding a query parameter to the browser URL.

IBM support might ask you to add trace logging for the UI, by updating the URL of your Event Endpoint Management UI:

https://some.eem.instance.com/catalog?logging=<VALUE>

When you navigate around the UI, it might seem that your logging configuration is not available. However, trace logging is enabled until you either close the UI browser page or click refresh.