Serving: Provide Docker for Mac integration test"

[jponge]

/kind bug

Expected Behavior

When I deploy a knative service and it scales down to 0, new pods should be respawned when issuing new requests.

Actual Behavior

Requests time out and no pod is being created.

Steps to Reproduce the Problem

1. Use Docker for Mac with Kubernetes
2. Install Knative as advised in https://github.com/knative/docs/blob/master/install/Knative-with-Docker-for-Mac.md
3. Deploy a knative serving hello world
4. Issue some requests to see it working
5. Wait until it scales down to 0
6. Issue new requests

[dgageot]

Same for me.

$ curl -v -H "Host: helloworld-go.default.example.com" http://localhost/
*   Trying ::1...
* TCP_NODELAY set
* Connection failed
* connect to ::1 port 80 failed: Connection refused
*   Trying 127.0.0.1...
* TCP_NODELAY set
* Connected to localhost (127.0.0.1) port 80 (#0)
> GET / HTTP/1.1
> Host: helloworld-go.default.example.com
> User-Agent: curl/7.54.0
> Accept: */*
>
< HTTP/1.1 504 Gateway Timeout
< content-length: 24
< content-type: text/plain
< date: Thu, 03 Jan 2019 11:54:14 GMT
< server: envoy
<
* Connection #0 to host localhost left intact
upstream request timeout

$ kubectl get pods
No resources found.

$ kubectl get ksvc -o yaml
apiVersion: v1
items:
- apiVersion: serving.knative.dev/v1alpha1
  kind: Service
  metadata:
    annotations:
      kubectl.kubernetes.io/last-applied-configuration: |
        {"apiVersion":"serving.knative.dev/v1alpha1","kind":"Service","metadata":{"annotations":{},"name":"helloworld-go","namespace":"default"},"spec":{"runLatest":{"configuration":{"revisionTemplate":{"spec":{"container":{"env":[{"name":"TARGET","value":"Go Sample v1"}],"image":"dgageot/helloworld-go"}}}}}}}
    clusterName: ""
    creationTimestamp: "2019-01-03T11:41:36Z"
    generation: 1
    name: helloworld-go
    namespace: default
    resourceVersion: "3927"
    selfLink: /apis/serving.knative.dev/v1alpha1/namespaces/default/services/helloworld-go
    uid: 869ea79a-0f4c-11e9-abdd-025000000001
  spec:
    generation: 1
    runLatest:
      configuration:
        revisionTemplate:
          metadata:
            creationTimestamp: null
          spec:
            container:
              env:
              - name: TARGET
                value: Go Sample v1
              image: dgageot/helloworld-go
              name: ""
              resources: {}
  status:
    address:
      hostname: helloworld-go.default.svc.cluster.local
    conditions:
    - lastTransitionTime: "2019-01-03T11:47:50Z"
      message: 'Revision "helloworld-go-00001" failed with message: "The target is
        not receiving traffic.".'
      reason: RevisionFailed
      status: "False"
      type: ConfigurationsReady
    - lastTransitionTime: "2019-01-03T11:47:50Z"
      message: 'Revision "helloworld-go-00001" failed with message: "The target is
        not receiving traffic.".'
      reason: RevisionFailed
      status: "False"
      type: Ready
    - lastTransitionTime: "2019-01-03T11:43:13Z"
      status: "True"
      type: RoutesReady
    domain: helloworld-go.default.example.com
    domainInternal: helloworld-go.default.svc.cluster.local
    latestCreatedRevisionName: helloworld-go-00001
    latestReadyRevisionName: helloworld-go-00001
    observedGeneration: 1
    traffic:
    - percent: 100
      revisionName: helloworld-go-00001
kind: List
metadata:
  resourceVersion: ""
  selfLink: ""

[markusthoemmes]

Is this using a released version or HEAD?

The instructions worked for v0.2.x. The newest version might need an update since the ingressgateway stuff has changed a little.

[dgageot]

For me, this is with v0.2.3

[jponge]

Same here, 0.2.3

[mattmoor]

/area networking

[tcnghia]

@dgageot is it possible for you to try removing istio-ingressgateway service and try again? may be this is the same issue as https://github.com/knative/serving/issues/2835 .

[dgageot]

@tcnghia I think you got the issue number wrong. I'll try again without the istio-ingressgateway

[jponge]

I didn't see this issue in 0.3+ so that was likely a bug in knative/serving.

[tcnghia]

@jponge are you talking about Knative 0.3+? So this is working for you now?

@dgageot can you please try with 0.3.x?

[jponge]

I had upgraded to 0.3.x and the auto-scaler issue had disappeared.

That being said, isn't there some integration test to catch this error?

On Thu, Jan 31, 2019 at 1:57 AM Nghia Tran [e-mail] wrote:

@jponge https://github.com/jponge are you talking about Knative 0.3+?
So this is working for you now?

@dgageot https://github.com/dgageot can you please try with 0.3.x?

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/knative/serving/issues/2835#issuecomment-459172292,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AABlaUzUQnKtu5a9PJXPFWHXQ_aiKqA1ks5vIj98gaJpZM4Zn2uw
.

[tcnghia]

Docker for Mac is one of the environments that is difficult for us to have integration tests on (probably need to run on Mac server or someone's MacBook?), but testing on multiple K8s providers would be nice.

Currently we only have integration tests running on GKE. I think Minikube may be another that we could run tests on.

/cc @adrcunha

[mattmoor]

I was asking about Docker-for-Mac problems in slack yesterday and @markusthoemmes indicated that he is using it without issue.

[mattmoor]

It also sounds like 0.3 resolved things, so I'm going to relabel this as a test-infrastructure feature request, but the Mac hardware requirement is going to make this tricky.

[markusthoemmes]

@mattmoor can you rename this to something like "Provide Docker for Mac integration test" or something?

[knative-housekeeping-robot]

Issues go stale after 90 days of inactivity.
Mark the issue as fresh by adding the comment /remove-lifecycle stale.
Stale issues rot after an additional 30 days of inactivity and eventually close.
If this issue is safe to close now please do so by adding the comment /close.

Send feedback to Knative Productivity Slack channel or file an issue in knative/test-infra.

/lifecycle stale