Troubleshooting a Knative Installation

This page describes how you can troubleshoot a Knative installation. Please try these instructions before filing a bug report.

Check if all containers are running, ready and healthy

Note

The outputs below are just examples. The containers may vary depending on the installation and configuration. Please make sure all containers are running, ready and healthy (look for the 1/1 in the following examples) or are Complete (in case of jobs).

Knative Serving Components

kubectl get pods -n knative-serving

NAME                                      READY   STATUS    RESTARTS   AGE
activator-6b9dc4c9db-cl56b                1/1     Running   0          2m
autoscaler-77f9b75856-f88qw               1/1     Running   0          2m
controller-7dcb56fdb6-dbzrp               1/1     Running   0          2m
domain-mapping-6bb8f95654-c575d           1/1     Running   0          2m
domainmapping-webhook-c77dcfcfb-hg2wv     1/1     Running   0          2m
webhook-78dc6ddddb-6868n                  1/1     Running   0          2m

Knative Serving Networking Layer

KourierIstioContour

kubectl get pods -n knative-serving

NAME                                      READY   STATUS    RESTARTS   AGE
net-kourier-controller-5fcbb6d996-fprpd   1/1     Running   0          103s

kubectl get pods -n kourier-system
NAME                                      READY   STATUS    RESTARTS   AGE
3scale-kourier-gateway-86b9f6dc44-xpn6h   1/1     Running   0          2m22s

kubectl get pods -n knative-serving

NAME                                    READY   STATUS    RESTARTS   AGE
net-istio-controller-ccc455b58-f98ld    1/1     Running   0          19s
net-istio-webhook-7558dbfc64-5jmt6      1/1     Running   0          19s

kubectl get pods -n istio-system

NAME                                   READY   STATUS    RESTARTS   AGE
istio-ingressgateway-c7b9f6477-bgr6q   1/1     Running   0          44s
istiod-79d65bf5f4-5zvtj                1/1     Running   0          29s

kubectl get pods -n knative-serving

NAME                                      READY   STATUS        RESTARTS   AGE
net-contour-controller-68547b797c-dl8pf   1/1     Running       0          14s

kubectl get pods -n contour-external

NAME                            READY   STATUS      RESTARTS   AGE
contour-7b995cdb68-jg5s8        1/1     Running     0          41s
contour-certgen-v1.24.2-zmr9r   0/1     Completed   0          41s
envoy-xkzck                     2/2     Running     0          41s

kubectl get pods -n contour-internal

NAME                            READY   STATUS      RESTARTS   AGE
contour-57fcf576fd-wb57c        1/1     Running     0          55s
contour-certgen-v1.24.2-gqgrx   0/1     Completed   0          55s
envoy-rht69                     2/2     Running     0          55s

Knative Eventing

kubectl get pods -n knative-eventing

NAME                                  READY   STATUS    RESTARTS   AGE
eventing-controller-bb8b689c4-lk6pq   1/1     Running   0          41s
eventing-webhook-577bb88ccd-hcz5p     1/1     Running   0          41s

Check if there are any errors logged in the Knative components

kubectl logs -n knative-serving <pod-name>
kubectl logs -n knative-eventing <pod-name>
kubectl logs -n <ingress-namespaces> <pod-namespaces> # see above for the relevant namespaces

# For example
kubectl logs -n knative-serving activator-6b9dc4c9db-cl56b
2023/05/01 11:52:51 Registering 3 clients
2023/05/01 11:52:51 Registering 3 informer factories
2023/05/01 11:52:51 Registering 4 informers

Check the status of the Knative Resources

Knative Serving

kubectl describe -n <namespace> kservice
kubectl describe -n <namespace> config
kubectl describe -n <namespace> revision
kubectl describe -n <namespace> sks # Serverless Service
kubectl describe -n <namespace> kingress # Knative Ingress
kubectl describe -n <namespace> rt # Knative Route
kubectl describe -n <namespace> dm # Domain-Mapping

# Check the status at the end. For example
kubectl describe -n default kservice

... omitted ...
Status:
  Address:
    URL:  http://hello.default.svc.cluster.local
  Conditions:
    Last Transition Time:        2023-05-01T12:08:18Z
    Status:                      True
    Type:                        ConfigurationsReady
    Last Transition Time:        2023-05-01T12:08:18Z
    Status:                      True
    Type:                        Ready
    Last Transition Time:        2023-05-01T12:08:18Z
    Status:                      True
    Type:                        RoutesReady
  Latest Created Revision Name:  hello-00001
  Latest Ready Revision Name:    hello-00001
  Observed Generation:           1
  Traffic:
    Latest Revision:  true
    Percent:          100
    Revision Name:    hello-00001
  URL:                http://hello.default.10.89.0.200.sslip.io
Events:
  Type    Reason   Age   From                Message
  ----    ------   ----  ----                -------
  Normal  Created  45s   service-controller  Created Configuration "hello"
  Normal  Created  45s   service-controller  Created Route "hello"

Knative Eventing

kubectl describe -n <namespace> brokers
kubectl describe -n <namespace> eventtypes
kubectl describe -n <namespace> triggers
kubectl describe -n <namespace> channels
kubectl describe -n <namespace> subscriptions
kubectl describe -n <namespace> apiserversources
kubectl describe -n <namespace> containersources
kubectl describe -n <namespace> pingsources
kubectl describe -n <namespace> sinkbindings


# Check the status at the end. For example
kubectl describe -n default brokers

... omitted ...
Status:
  Annotations:
    bootstrap.servers:                 my-cluster-kafka-bootstrap.kafka:9092
    default.topic.partitions:          10
    default.topic.replication.factor:  3