Enable docker devfile push test in OpenShift CI

[amitkrout]

What type of PR is this?

/kind feature

What does does this PR do / why we need it:
Enable docker devfile push test in OpenShift CI

Which issue(s) this PR fixes:

Fixes NA

How to test changes / Special notes to the reviewer:
docker devfile push test should pass in OpenShift CI

[openshift-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
To complete the pull request process, please assign amitkrout
You can assign the PR to them by writing /assign @amitkrout in a comment when ready.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• openshift-ci/build-root/OWNERS
• scripts/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[amitkrout]

#2806 (comment) cc_ @johnmcollier

[openshift-ci-robot]

@amitkrout: The following tests failed, say /retest to rerun all failed tests:

Test name	Commit	Details	Rerun command
ci/prow/v4.3-integration-e2e-benchmark	51c02dd	link	/test v4.3-integration-e2e-benchmark
ci/prow/v4.4-integration-e2e-benchmark	51c02dd	link	/test v4.4-integration-e2e-benchmark
ci/prow/v4.2-integration-e2e-benchmark	51c02dd	link	/test v4.2-integration-e2e-benchmark

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.

[girishramnani]

@amitkrout there is a legitimate failures on openshift CI. can you please debug this?
cc @maysunfaisal just in case you need help

[johnmcollier]

The tests are failing because there isn't a Docker socket available on the OpenShift CI:

[odo]  ✗  Failed to start component with name nflpne.
[odo] Error: Failed to create the component: unable to create or update component: Unable to create project source volume for component nflpne: error creating docker volume: Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?
Removing locally running Docker images with label component=nflpneListing locally running Docker images with label component=nflpneRunning docker with args [docker ps -q --filter label=component=nflpne]

In order to run the Docker tests there, we'd need to find a way to set up a remote docker daemon (maybe running in another container?) and get odo/docker to talk to it instead of /var/run/docker.sock

But I'm wondering, do we need the Docker tests to run on the OpenShift CI? Since they won't be interacting with a Kube environment, we shouldn't need coverage on different OCPs. The Travis CI should be sufficient for test coverage for odo with Docker.

[kadel]

Enable docker devfile push test in OpenShift CI

Can we actually do this? Test on Prow run inside Pod, I'm not sure if you will be able to access docker daemon inside container on OpenShift

[kadel]

But I'm wondering, do we need the Docker tests to run on the OpenShift CI? Since they won't be interacting with a Kube environment, we shouldn't need coverage on different OCPs. The Travis CI should be sufficient for test coverage for odo with Docker.

+1

[amitkrout]

@johnmcollier @kadel I just opend the pr just to have more discussoin on it

The tests are failing because there isn't a Docker socket available on the OpenShift CI:
[odo] ✗ Failed to start component with name nflpne.
[odo] Error: Failed to create the component: unable to create or update component: Unable to create project source volume for component nflpne: error creating docker volume: Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?
Removing locally running Docker images with label component=nflpneListing locally running Docker images with label component=nflpneRunning docker with args [docker ps -q --filter label=component=nflpne]

@johnmcollier The error is coming because of docker is not started in the test container. Here is the log details - https://storage.googleapis.com/origin-ci-test/pr-logs/pull/openshift_odo/2921/pull-ci-openshift-odo-master-v4.2-integration-e2e-benchmark/2026/artifacts/build-logs/root.log.

Ideally the docker demon should start as per the change i made https://github.com/openshift/odo/pull/2921/files#diff-01f4ec918b634e024905add2f91a58abR5-R6.

But I'm wondering, do we need the Docker tests to run on the OpenShift CI? Since they won't be interacting with a Kube environment, we shouldn't need coverage on different OCPs. The Travis CI should be sufficient for test coverage for odo with Docker.

I miss interpret the whole test scenario, actually the line https://github.com/openshift/odo/blob/master/.travis.yml#L150 force me to test it against 4.* too. Now understood that for docker push target, no OpenShift cluster is needed. Thanks @johnmcollier.

So instead of oc version i think we should add coverage for docker version. I am creating a separate issue #2995 for the same.

Can we actually do this? Test on Prow run inside Pod, I'm not sure if you will be able to access docker daemon inside container on OpenShift

@kadel I think there should not be any issues as we are triggering the test inside a container. We just need to setup a docker environment inside that container. Anyway it will add no sense as per @johnmcollier statement.