search

openshift-pipelines / pipeline-service

SaaS for Tekton Pipelines
Apache License 2.0
23 stars 44 forks source link

tektoncd-results update #957

Closed rh-tap-build-team[bot] closed 6 months ago

rh-tap-build-team[bot] commented 6 months ago

Included PRs:

enarha commented 6 months ago

/retest

gabemontero commented 6 months ago

/retest

gabemontero commented 6 months ago

so far these have been env flakes prior to pipeline svc getting deployed

if any of these flakes repeat with any frequency I'll reach on to Xin in slack

gabemontero commented 6 months ago

I think the docker image failure is related to dependency tagging @Roming22 noted in the update binaries PR, though I've asked him in slack

The deploy-cluster error in the vanilla test deployment feels either like a flake or a path it has gone down where it is attempting a kubectl get.. against some of the Rosa et. al. stuff which is not allowed, but I've asked in slack including pinging @xinredhat about that

given the upgrade test passed, I'm leaning toward that being sufficient validation for merging this, but will wait until I get some responses either here or in slack about my questions

gabemontero commented 6 months ago

/retest

gabemontero commented 6 months ago

rosa fail this time around on deployment/upgrade

image fail is the git version again

/retest

gabemontero commented 6 months ago

/retest

Roming22 commented 6 months ago

/retest

gabemontero commented 6 months ago

So @Roming22 I did see one instance of I believe the upgrade test passing before the cluster deploy issues @xinredhat has opened trackers for took hold.

I'm fine with waiting some, but if the ^^ situation ends up taking a while to resolve, I'd like to entertain merging this with that prior test passage, plus perhaps some manual verification where I dev_setup this branch and do the results testing locally.

Results performance has popped up a few times now in prod. Getting relief for the leak so we can start the path of increasing the worker thread pool is hence rising in priority.

WDYT?

gabemontero commented 6 months ago

OK I checked out this PR's branch into https://github.com/gabemontero/pipeline-service/tree/local-upstream-957-copy and then ran ./developer/openshift/dev_setup.sh --use-current-branch and tested locally.

I'm merging this, given the known rosa/hypershift CI issues, with these validations, and we'll move to the infra-deployment bumps, where must gather in any e2e's will let me also confirm results/logs storage and pruning