Pipeline: TestTaskRunRetry is flaky

Created on 28 Jul 2020  路  5Comments  路  Source: tektoncd/pipeline

TestTaskRunRetry is flaky.

I run into this several times for a PR today and for another PR yesterday.

examples:

/kind flake

aretesting help wanted kinflake
Source
picture jlpettersson

All 5 comments

/assign @thomaschandler

thomaschandler picture thomaschandler on 29 Jul 2020
🎉1

@jlpettersson I wonder if this might be related to the garbage collection of failed pods. I'm unsure if GKE has a different garbage collection process than a self hosted k8s setup.

I'm thinking of updating TestTaskRunRetry to wait for pod state changes, at which point it will record the pod information for the failed pod, which can later be asserted once the PipelineRun is complete. Think this is worth a shot?

thomaschandler picture thomaschandler on 6 Aug 2020
👍1

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

/lifecycle stale

Send feedback to tektoncd/plumbing.

tekton-robot picture tekton-robot on 4 Nov 2020

This is happening again/still: https://tekton-releases.appspot.com/build/tekton-prow/pr-logs/pull/tektoncd_pipeline/3494/pull-tekton-pipeline-integration-tests/1324621677065146370/


TestTaskRunRetry: retry_test.go:119: BUG: Found 5 Pods, want 6
ImJasonH picture ImJasonH on 6 Nov 2020

/remove-lifecycle stale

ImJasonH picture ImJasonH on 6 Nov 2020
Was this page helpful?
0 / 5 - 0 ratings

Related issues

castlemilk picture castlemilk  路  4Comments
objectiveous picture objectiveous  路  3Comments
bobcatfish picture bobcatfish  路  4Comments
sbwsg picture sbwsg  路  3Comments
VeereshAradhya picture VeereshAradhya  路  3Comments