now served with tensorflow/serving, it will accept requests using 'inputs'
    and 'outputs'. Starting at 1.2, such a model will accept the keys specified
    during export. Therefore, inference requests using 'inputs' and 'outputs'
    may start to fail. To fix this, either update any inference clients to send
    requests with the actual input and output keys used by the trainer code, or

	reqInfo := logger.GetReqInfo(ctx)
	var newReqInfo *logger.ReqInfo
	if reqInfo != nil {
		newReqInfo = logger.NewReqInfo(reqInfo.RemoteHost, reqInfo.UserAgent, reqInfo.DeploymentID, reqInfo.RequestID, reqInfo.API, bucket, object)
	} else {
		newReqInfo = logger.NewReqInfo("", "", globalDeploymentID(), "", "Heal", bucket, object)
	}
	healCtx := logger.SetReqInfo(GlobalContext, newReqInfo)

	var lastReq *http.Request
	checkRedirect := func(req *http.Request, via []*http.Request) error {
		// fmt.Printf("CheckRedirect:\n")
		// fmt.Printf("Upcoming: %s %s\n", req.Method, req.URL.String())
		// for i, c := range via {
		// 	fmt.Printf("Sofar %d: %s %s\n", i, c.Method, c.URL.String())
		// }
		// Save the last request in a redirect chain.
		lastReq = req

In versions lower than `0.65.2`, FastAPI would try to read the request payload as JSON even if the `content-type` header sent was not set to `application/json` or a compatible JSON media type (e.g. `application/geo+json`).

So, a request with a content type of `text/plain` containing JSON data would be accepted and the JSON data would be extracted.

* Add `metrics-port` to kube-proxy cmd flags. ([#72682](https://github.com/kubernetes/kubernetes/pull/72682), [@whypro](https://github.com/whypro))