WIP: TESTING: DO NOT MERGE

[cevich]

Signed-off-by: Chris Evich [email protected]

[openshift-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: cevich
To complete the pull request process, please assign rhatdan
You can assign the PR to them by writing /assign @rhatdan 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:

• OWNERS

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

[cevich]

@edsantiago I think the spot you're seeing the "hang" is a red-herring, I think something's happening on another level to cause this. Look at the example from this PR: https://cirrus-ci.com/task/5905973485240320?command=system_test#L156 Execution simply/literally stops w/o notice, until the "BIG" timeout clobbers it. All the other tests passed, so...geeze...the only thing I can think is something must be happening on the cirrus side. Maybe related to the underlying VMs being preemptable and a secondary bug related to output processing when that happens? https://cloud.google.com/compute/docs/instances/preemptible

[cevich]

....Actually, no...scratch that. Our VMs are NOT preemptable...we use the expensive ones 😕

[edsantiago]

FWIW: acting on the possibility that the env bash PR could be responsible, I rebased my two PRs on the commit prior to that. No luck on either one: both ran into the 60- and 120-minute timeout, same tests, same symptoms.

[cevich]

FWIW: acting on the possibility that the env bash PR could be responsible

Excellent, thanks for trying that. One more question-mark item we can check off the list.

ONE of my tests also hit the 120min timeout: https://github.com/containers/podman/pull/7342/checks?check_run_id=995072852 but appears almost totally different from yours. I'm taking this as meaning the problem isn't specific to your PR, and we should basically expect to see this problem appear everywhere.

I'm going to try and re-run it to see if the behavior changes at all (IIRC, yours always failed for at least 3-5 retries).

[cevich]

Problem fixed by #7354