stopFn := globalBatchJobsMetrics.trace(batchJobMetricReplication, job.ID, attempts)
				success := true
				if err := r.ReplicateFromSource(ctx, api, core, oi, retry); err != nil {
					// object must be deleted concurrently, allow these failures but do not count them
					if isErrVersionNotFound(err) || isErrObjectNotFound(err) {
						return
					}
					stopFn(oi, err)
					batchLogIf(ctx, err)
					success = false

	return writeAllMetadataWithRevert(ctx, disks, origbucket, bucket, prefix, files, quorum, true)
}

// writeUniqueFileInfo - writes unique `xl.meta` content for each disk concurrently.
func writeUniqueFileInfo(ctx context.Context, disks []StorageAPI, origbucket, bucket, prefix string, files []FileInfo, quorum int) ([]StorageAPI, error) {

	// The time of the last scan/heal activity
	lastHealActivity time.Time

	// Holds the request-info for logging
	ctx context.Context

	// used to lock this structure as it is concurrently accessed
	mutex sync.RWMutex
}

// NewHealSequence - creates healSettings, assumes bucket and
// objPrefix are already validated.
func newHealSequence(ctx context.Context, bucket, objPrefix, clientAddr string,

  public final Optional<@NonNull E> last() {
    // Iterables#getLast was inlined here so we don't have to throw/catch a NSEE

    // TODO(kevinb): Support a concurrently modified collection?
    Iterable<E> iterable = getDelegate();
    if (iterable instanceof List) {
      List<E> list = (List<E>) iterable;
      if (list.isEmpty()) {
        return Optional.absent();
      }

* Others...

## Replication - Processes and Memory

With a FastAPI application, using a server program like the `fastapi` command that runs Uvicorn, running it once in **one process** can serve multiple clients concurrently.

But in many cases, you will want to run several worker processes at the same time.

### Multiple Processes - Workers

 * methods for more details.
 *
 * <p>Note that this implementation is not synchronized. If multiple threads access the same cell of
 * an {@code ArrayTable} concurrently and one of the threads modifies its value, there is no
 * guarantee that the new value will be fully visible to the other threads. To guarantee that

    strictly ASCII. This issue is tracked as [CVE-2021-0341].


## Version 4.9.1

_2021-01-30_

 *  Fix: Work around a crash in Android 10 and 11 that may be triggered when two threads
    concurrently close an SSL socket. This would have appeared in crash logs as
    `NullPointerException: bio == null`.


## Version 4.9.0

_2020-09-11_

 *  New: Log the TLS handshake in `LoggingEventListener`.

## Version 3.12.13

_2021-01-30_

 *  Fix: Work around a crash in Android 10 and 11 that may be triggered when two threads
    concurrently close an SSL socket. This would have appeared in crash logs as
    `NullPointerException: bio == null`.


## Version 3.12.12

_2020-05-17_

import java.util.NoSuchElementException;
import java.util.Set;
import java.util.concurrent.CancellationException;
import java.util.concurrent.ConcurrentMap;
import java.util.concurrent.atomic.AtomicInteger;
import java.util.concurrent.atomic.AtomicReferenceArray;
import java.util.concurrent.locks.ReentrantLock;
import javax.annotation.CheckForNull;

/**
 * The concurrent hash map implementation built by {@link MapMaker}.
 *

   * @throws NoSuchElementException if the iterable is empty
   */
  @ParametricNullness
  public static <T extends @Nullable Object> T getLast(Iterable<T> iterable) {
    // TODO(kevinb): Support a concurrently modified collection?
    if (iterable instanceof List) {
      List<T> list = (List<T>) iterable;
      if (list.isEmpty()) {
        throw new NoSuchElementException();
      }