* MockWebServer. In cases where logic is particularly tricky, we use unit tests. This class makes
 * it easy to get sample values to use in such tests.
 *
 * This class is pretty fast and loose with default values: it attempts to provide values that are
 * well-formed, but doesn't guarantee values are internally consistent. Callers must take care to
 * configure the factory when sample values impact the correctness of the test.

                resumeKey = rkresp.getResumeKey();

                // start with some reasonably safe defaults, the server will till us if it does not like it
                // can we resume this if we loose the file descriptor?

                int maxChunks = 256;
                int maxChunkSize = 1024 * 1024;
                int byteLimit = 16 * 1024 * 1024;
                boolean retry = false;
                do {

	if err != nil {
		// Failed to create newbucket, abort.
		t.Fatalf("%s : %s", instanceType, err.Error())
	}

	// Take 4 disks down, one more we loose quorum on 16 disk node.
	for _, disk := range disks[:4] {
		os.RemoveAll(disk)
	}

	testCases := []struct {
		bucketName    string
		objName       string
		inputData     []byte

be <i>exact</i> or <i>loose</i>.
As unification recursively descends a composite type structure,
the matching mode used for elements of the type, the <i>element matching mode</i>,
remains the same as the matching mode except when two types are unified for
<a href="#Assignability">assignability</a> (<code>≡<sub>A</sub></code>):
in this case, the matching mode is <i>loose</i> at the top level but

    public HashCode hash() {
      long value = checksum.getValue();
      if (bits == 32) {
        /*
         * The long returned from a 32-bit Checksum will have all 0s for its second word, so the
         * cast won't lose any information and is necessary to return a HashCode of the correct
         * size.
         */
        return HashCode.fromInt((int) value);
      } else {
        return HashCode.fromLong(value);
      }
    }

### Checking cluster health for maintenance

You may query the cluster probe endpoint to check if the node which received the request can be taken down for maintenance, if the server replies back '412 Precondition Failed' this means you will lose HA. '200 OK' means you are okay to proceed.

```
curl http://minio1:9001/minio/health/cluster?maintenance=true
HTTP/1.1 412 Precondition Failed
Accept-Ranges: bytes
Content-Length: 0
Server: MinIO
Vary: Origin

## Why is Erasure Code useful?

	_, err = obj.PutObject(ctx, bucket, object, mustGetPutObjReader(t, bytes.NewReader([]byte("abcd")), int64(len("abcd")), "", ""), opts)
	if err != nil {
		t.Fatal(err)
	}

	// Remove one more disk to 'lose' quorum, by taking 2 more drives offline.
	erasureDisks = xl.getDisks()
	z.serverPools[0].erasureDisksMu.Lock()
	xl.getDisks = func() []StorageAPI {
		erasureDisks[7] = nil
		erasureDisks[8] = nil

All of the cashiers doing all the work with one client after the other 👨‍💼⏯.

And you have to wait 🕙 in the line for a long time or you lose your turn.

You probably wouldn't want to take your crush 😍 with you to run errands at the bank 🏦.

### Burger Conclusion { #burger-conclusion }

 *  Fix: Handshake now returns peer certificates in canonical order: each certificate is signed by
    the certificate that follows and the last certificate is signed by a trusted root.

 *  Fix: Don't lose HTTP/2 flow control bytes when incoming data races with a stream close. If this
    happened enough then eventually the connection would stall.