### Restart After Crash

But in those cases with really bad errors that crash the running **process**, you would want an external component that is in charge of **restarting** the process, at least a couple of times...

!!! tip

type PartTooSmall struct {
	PartSize   int64
	PartNumber int
	PartETag   string
}

func (e PartTooSmall) Error() string {
	return fmt.Sprintf("Part size for %d should be at least 5MB", e.PartNumber)
}

// PartTooBig returned if size of part is bigger than the allowed limit.
type PartTooBig struct{}

func (e PartTooBig) Error() string {
	return "Part size bigger than the allowed limit"

		// Run the data scanner in a loop
		for {
			runDataScanner(ctx, objAPI)
			duration := time.Duration(r.Float64() * float64(scannerCycle.Load()))
			if duration < time.Second {
				// Make sure to sleep at least a second to avoid high CPU ticks.
				duration = time.Second
			}
			time.Sleep(duration)
		}
	}()
}

func getCycleScanMode(currentCycle, bitrotStartCycle uint64, bitrotStartTime time.Time) madmin.HealScanMode {

	_, err := client.Kube().CoreV1().Namespaces().Patch(ctx, ns.Name,
		types.MergePatchType, labelsPatch, metav1.PatchOptions{})
	assert.NoError(t, err)

	// wait until at least one add event happens
	mt.Assert(EventTotals.Name(), map[string]string{"type": "add"}, monitortest.AtLeast(1))

	assertPodNotAnnotated(t, client, pod)

	// Assert expected calls (not) actually made

	// prefers setCounts to be sorted for optimal behavior.
	if divisibleSize < setCounts[len(setCounts)-1] {
		return divisibleSize
	}

	// Figure out largest value of total_drives_in_erasure_set which results
	// in least number of total_drives/total_drives_erasure_set ratio.
	prevD := divisibleSize / setCounts[0]
	for _, cnt := range setCounts {
		if divisibleSize%cnt == 0 {
			d := divisibleSize / cnt
			if d <= prevD {
				prevD = d

Docker and similar tools also use an **internal cache** when building the image, if a file hasn't changed since the last time building the container image, then it will **re-use the same layer** created the last time, instead of copying the file again and creating a new layer from scratch.

| `minio_cluster_iam_plugin_authn_service_last_succ_seconds`      | `counter` | When plugin authentication is configured, returns time (in seconds) since the last successful request to the service     |        |

apiVersion: v1
kind: ServiceAccount
metadata:
  name: {{.ServiceAccount | quote}}
  namespace: {{.Namespace | quote}}
  annotations:
    {{- toJsonMap (omit .InfrastructureAnnotations "kubectl.kubernetes.io/last-applied-configuration" "gateway.istio.io/name-override" "gateway.istio.io/service-account" "gateway.istio.io/controller-version") | nindent 4 }}
  labels:
    {{- toJsonMap
      .InfrastructureLabels
      (strdict

The CNI plugins are executed by threads in the `kubelet` process.  The CNI plugins logs end up the syslog
under the `kubelet` process. On systems with `journalctl` the following is an example command line
to view the last 1000 `kubelet` logs via the `less` utility to allow for `vi`-style searching:

```console
$ journalctl -t kubelet -n 1000 | less
```

#### GKE via Stackdriver Log Viewer

		// Only replace if needed.
		// Existing requests will use the previous limit,
		// but new requests will use the new limit.
		// There will be a short overlap window,
		// but this shouldn't last long.
		t.requestsPool = make(chan struct{}, apiRequestsMaxPerNode)
	}
	t.requestsDeadline = cfg.RequestsDeadline
	listQuorum := cfg.ListQuorum
	if listQuorum == "" {
		listQuorum = "strict"
	}