bump test e2e timeout to 100m

[mkenigs]

e2e-gcp-op has been failing on a lot of PRs, and while continuing to
increase the timeout long term is not the solution, it seems like
bumping the timeout for now is better than mashing retest until we fix
it

[kikisdeliveryservice]

Why is the test timing out? Is this a functional issue or were other tests added? If the latter there might be a bug. When did this start?
Til further info:
/hold

[mkenigs]

Not sure. At least as early as #2976 (comment)

[kikisdeliveryservice]

Not sure. At least as early as #2976 (comment)

I think we're more interested in when it first started - was it a PR or did something else change?

[openshift-ci]

@mkenigs: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
ci/prow/e2e-aws	ba532ba	link	true	/test e2e-aws

Full PR test history. Your PR dashboard.

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.

[cgwalters]

I agree it'd be good to investigate, but I think merging this PR as is is better than the status quo of wasting CI resources on retesting until PRs eventually make it in.

[cgwalters]

/approve
Just from my PoV, but not doing lgtm or lifting hold, that's team consensus.

This PR was 85.6 minutes for reference. But yeah, looking at e.g. this job from the release-4.10 branch the run as at 78 minutes.

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: cgwalters, mkenigs

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

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [cgwalters]

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

[cgwalters]

Just eyeballing it, some differences:

4.11: --- PASS: TestKubeletConfigMaxPods (650.27s)
4.10: --- PASS: TestKubeletConfigMaxPods (480.11s)

4.11: --- PASS: TestKernelArguments (489.49s)
4.10: --- PASS: TestKernelArguments (295.44s)

Computing how much variance there is here and digging into why is another task...

[kikisdeliveryservice]

gcp-op is mainly green so it's not really blocking merges. We have before just extended test times and later on realized there was indeed a bug. Since this isn't blocking prs from merging (other jobs are doing that) would prefer to know when this started and root cause instead of just extending time.

Those times Colin points out above are very very not trivial. =/