The work for this feature was already started, but never merged.

This work is for classic load balancers only. See CFEPLAN-54 for details.

Tasks:

• Reopen PRs

• • https://github.com/openshift/enhancements/pull/461

• • https://github.com/openshift/api/pull/730

• • https://github.com/openshift/cluster-ingress-operator/pull/451

• Rebase PRs
• Review and merge PRs
• Add tests (unit and/or e2e)

Acceptance criteria:

• When I create an IngressController that specifies a short timeout (for example, 10 seconds), with a route to a pod that serves HTTP requests with static content, I expect the connection to timeout after 10s if the connection is open but no traffic is sent.
• When I create an IngressController that specifies a longer timeout (for example, 2min), with a route to a pod that serves HTTP requests with static content, I expect a HTTP200 response to be received if the connection is open, but the request is sent more than a minute after that.
•