disconnect openapi aggregation

[deads2k]

Fixes #15326

This does the minimal diff to disconnect the openapi aggregation to make future picks/fixes easier to bring back.

@liggitt @sttts @smarterclayton ptal.

[deads2k]

hopefully this is a little faster.

[liggitt]

/lgtm

[openshift-merge-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: deads2k, liggitt

No associated issue. Update pull-request body to add a reference to an issue, or get approval with /approve no-issue

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

Needs approval from an approver in each of these OWNERS Files:

• OWNERS [deads2k,liggitt]

You can indicate your approval by writing /approve in a comment
You can cancel your approval by writing /approve cancel in a comment

[deads2k]

This is expected to significantly decrease test-integration time. Bumping to the top of the queue.

[liggitt]

took 20 minutes off the time

[smarterclayton]

You'll like this.

Before: one integration test took about 40 wall seconds and about 38 core seconds.
After: one integration test took about 20 wall seconds and 0.8 core seconds

[smarterclayton]

Of the remaining 0.8 core seconds.... 0.6 core seconds are directly attributable to openapi parsing on startup.

[deads2k]

Before: one integration test took about 40 wall seconds and about 38 core seconds.
After: one integration test took about 20 wall seconds and 0.8 core seconds

But lacked sufficient imagination. It's taking me something like six seconds to have openapi start even with this pull. I think this should still go in, but I'll keep chasing it. We don't need openapi for integration tests.

[sttts]

Hou, what a difference. What made it that slow?

[deads2k]

I don't see any swagger changes between the test and now. letting it through without a retest

[smarterclayton]

I think I might have found the 20 characters that make all of this go away :)

[openshift-merge-robot]

Automatic merge from submit-queue