// ConnSettings - contains connection settings.
type ConnSettings struct {
	DialContext DialContext // Custom dialContext, DialTimeout is ignored if this is already setup.
	LookupHost  LookupHost  // Custom lookupHost, is nil on containerized deployments.
	DialTimeout time.Duration

	// TLS Settings
	RootCAs          *x509.CertPool
	CipherSuites     []uint16
	CurvePreferences []tls.CurveID

	// HTTP2
	EnableHTTP2 bool

	// TCP Options

## Pod Shutdown

### Pod Shutdown Requirements

During Pod shutdown, we MUST meet the constraint that Pod traffic will be handled until all containerized processes within the Pod have terminated.

The details for the deployment & installation of this plugin were pretty much lifted directly from the
[Calico CNI plugin](https://github.com/projectcalico/cni-plugin).

Specifically:

- The CNI installation script is containerized and deployed as a daemonset in k8s.  The relevant calico k8s manifests were used as the model for the istio-cni plugin's manifest:

								startTime.Add(timeElapsed),
								"elapsed",
								"",
							))
						ctx := logger.SetReqInfo(GlobalContext,
							reqInfo)
						bootLogOnceIf(ctx, fmt.Errorf("%s resolves to localhost in a containerized deployment, waiting for it to resolve to a valid IP",
							endpoints[i].Hostname()), endpoints[i].Hostname(), logger.ErrorKind)
					}

					continue
				}

				// return err if not Docker or Kubernetes

* fix nsenter GetFileType issue in containerized kubelet ([#62467](https://github.com/kubernetes/kubernetes/pull/62467), [@andyzhangx](https://github.com/andyzhangx))

- change default azure file mount permission to 0777 ([#69854](https://github.com/kubernetes/kubernetes/pull/69854), [@andyzhangx](https://github.com/andyzhangx))
- Fixed subpath in containerized kubelet. ([#69565](https://github.com/kubernetes/kubernetes/pull/69565), [@jsafrane](https://github.com/jsafrane))

	var podUID types.UID
	var containerID string
	for _, cgroup := range cgroups {
		candidatePodUID, candidateContainerID, ok := getPodUIDAndContainerIDFromCGroupPath(cgroup.GroupPath)
		switch {
		case !ok:
			// Cgroup did not contain a container ID.
			continue
		case containerID == "":
			// This is the first container ID found so far.
			podUID = candidatePodUID

		},
	} {
		t.Run(tt.name, func(t *testing.T) {
			podUID, containerID, err := getPodUIDAndContainerIDFromCGroups(makeCGroups(tt.cgroupPaths))

			if tt.expectMsg != "" {
				assert.Equal(t, tt.expectMsg, err.Error())
				return
			}
			assert.Equal(t, tt.expectPodUID, podUID)
			assert.Equal(t, tt.expectContainerID, containerID)
		})
	}

* ScaleIO driver completely removes dependency on drv_cfg binary so a Kubernetes cluster can easily run a containerized kubelet. ([#54956](https://github.com/kubernetes/kubernetes/pull/54956), [@vladimirvivien](https://github.com/vladimirvivien))



# v1.8.3

- Integration testing now takes periodic Prometheus scrapes from the etcd server.
  There is a new script ,`hack/run-prometheus-on-etcd-scrapes.sh`, that runs a containerized Prometheus server against an archive of such scrapes. ([#106190](https://github.com/kubernetes/kubernetes/pull/106190), [@MikeSpreitzer](https://github.com/MikeSpreitzer)) [SIG API Machinery and Testing]