Unexpected start times reported for skipped steps

[AlanGreene]

Expected Behavior

Step start times should be sequential regardless of whether they actually ran or not.

Actual Behavior

If a TaskRun fails, any remaining skipped steps retain their original container start time instead of being updated by the entrypoint binary.

Steps to Reproduce the Problem

kubectl create -f - << EOF
---
apiVersion: tekton.dev/v1beta1
kind: TaskRun
metadata:
  generateName: test-start-time-correct-
spec:
  taskSpec:
    steps:
    - image: ubuntu
      script: |
        #!/usr/bin/env bash
        sleep 3 && echo step 1
    - image: ubuntu
      script: |
        #!/usr/bin/env bash
        sleep 3 && echo step 2
    - image: ubuntu
      script: |
        #!/usr/bin/env bash
        sleep 3 && echo step 3
    - image: ubuntu
      script: |
        #!/usr/bin/env bash
        sleep 3 && echo step 4
---
apiVersion: tekton.dev/v1beta1
kind: TaskRun
metadata:
  generateName: test-start-time-wrong-
spec:
  taskSpec:
    steps:
    - image: ubuntu
      script: |
        #!/usr/bin/env bash
        sleep 3 && echo step 1
    - image: ubuntu
      script: |
        #!/usr/bin/env bash
        sleep 3 && echo step 2
        false
    - image: ubuntu
      script: |
        #!/usr/bin/env bash
        sleep 3 && echo step 3
    - image: ubuntu
      script: |
        #!/usr/bin/env bash
        sleep 3 && echo step 4
EOF

The step start times for test-start-time-correct- are recorded and reported correctly, reflecting the time the step actually executed.

The step start times for the first 2 steps in test-start-time-wrong- are correct, but the last 2 skipped steps still report their original container start times. This means that if I order the step status by start time they appear before the steps that actually ran.

Additional Info

• Kubernetes version:

  Output of kubectl version:

  Client Version: v1.16.3
  Server Version: v1.15.5
  

• Tekton Pipeline version:

  Output of tkn version or kubectl get pods -n tekton-pipelines -l app=tekton-pipelines-controller -o=jsonpath='{.items[0].metadata.labels.version}'

  v0.11.1

/kind bug

Related issue: #1512
This fixed the start times for steps that are actually executed.

[vincent-pli]

Fixed by #2429

[vdemeester]

/close

[tekton-robot]

@vdemeester: Closing this issue.

In response to this:

/close

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.

[AlanGreene]

Thanks @vincent-pli, verified on the latest nightly build.