warning=$$(cat manifests/helm-profiles/warning-edit.txt | sed ':a;N;$$!ba;s/\n/\\n/g') ; \
	for chart in $(CHARTS) ; do \
		for profile in manifests/helm-profiles/*.yaml ; do \
			sed "1s|^|$${warning}\n\n|" $$profile > manifests/charts/$$chart/files/profile-$$(basename $$profile) ; \
		done; \
		cp manifests/zzz_profile.yaml manifests/charts/$$chart/templates ; \
	done

    server.enqueue(
      MockResponse.Builder()
        .body("def")
        .throttleBody(1, 750, TimeUnit.MILLISECONDS)
        .build(),
    )

    // First request: time out after 1s.
    client =
      client.newBuilder()
        .readTimeout(Duration.ofSeconds(1))
        .build()
    executeSynchronously("/a").assertBody("abc")

    // Second request: time out after 250ms.

- Ensure Job status updates are batched by 1s. This fixes an unlikely scenario when a sequence of immediately 

- GRPC probes now set a linger option of 1s to improve the TIME-WAIT state. ([#115321](https://github.com/kubernetes/kubernetes/pull/115321), [@rphillips](https://github.com/rphillips)) [SIG Network and Node]