return
	}
	remote := m.targets[cReq.Host]
	if remote == nil {
		writeErr(fmt.Errorf("unknown incoming host: %v", cReq.Host))
		return
	}
	if time.Since(cReq.Time).Abs() > 5*time.Minute {
		writeErr(fmt.Errorf("time difference too large between servers: %v", time.Since(cReq.Time).Abs()))
		return
	}
	if err := m.authToken(cReq.Token); err != nil {
		writeErr(fmt.Errorf("auth token: %w", err))
		return
	}

Realistically, that means the best we can do is:
* Ensure that, at any point in time, any _new_ connections will succeed. There is no period where new connections are dropped.
* Provide some period of time for the old Ztunnel to continue processing any established connections.

If this period of time ("drain period") is longer than any connections, there is zero impact on applications at all.

   * e]}.
   */
  @Override
  String toString();

  // Refined Collection Methods

  /**
   * {@inheritDoc}
   *
   * <p>Elements that occur multiple times in the multiset will appear multiple times in this
   * iterator, though not necessarily sequentially.
   */
  @Override
  Iterator<E> iterator();

  /**
   * Determines whether this multiset contains the specified element.

pre-Go 1.24 behavior.

### Go 1.23

Go 1.23 changed the channels created by package time to be unbuffered
(synchronous), which makes correct use of the [`Timer.Stop`](/pkg/time/#Timer.Stop)
and [`Timer.Reset`](/pkg/time/#Timer.Reset) method results much easier.
The [`asynctimerchan` setting](/pkg/time/#NewTimer) disables this change.
There are no runtime metrics for this change,

			globalTrace.Publish(scannerTrace(s, startTime, duration, strings.Join(paths, " "), custom))
		}
	}
}

// time a scanner action.
// Use for s < scannerMetricLastRealtime
func (p *scannerMetrics) time(s scannerMetric) func() {
	startTime := time.Now()
	return func() {
		duration := time.Since(startTime)

		atomic.AddUint64(&p.operations[s], 1)
		if s < scannerMetricLastRealtime {
			p.latency[s].add(duration)

import com.google.j2objc.annotations.J2ObjCIncompatible;
import java.time.Duration;
import java.util.concurrent.TimeUnit;

/**
 * An object that accurately measures <i>elapsed time</i>: the measured duration between two
 * successive readings of "now" in the same process.
 *
 * <p>In contrast, <i>wall time</i> is a reading of "now" as given by a method like

   * timeouts (due to a problem with the transport). When a stream times out we don't know whether
   * the problem impacts just one stream or the entire connection.
   *
   * To differentiate the two cases we ping the server when a stream times out. If the overall
   * connection is fine the ping will receive a pong; otherwise it won't.
   *

							<c:if test="${eoled}">
								<li class="nav-item" data-toggle="tooltip" data-placement="left" title="<la:message key="labels.eol_error" />">
									<a class="nav-link active" href="${eolLink}" target="_olh"><em class="fas fa-times-circle text-danger"></a>
								</li>
							</c:if>
							<c:if test="${developmentMode}">
								<li class="nav-item" data-toggle="tooltip" data-placement="left"

ig.allowDuplicates&&(l+="-"+Math.floor(1e3*Math.random()),s+="-"+Math.floor(1e3*Math.random()));var r='<li class="nav-item" role="presentation"><a href="#" class="btn-iframe-close" data-widget="iframe-close" data-type="only-this"><i class="fas fa-times"></i></a><a class="nav-link" data-toggle="row" id="'+s+'" href="#'+l+'" role="tab" aria-controls="'+l+'" aria-selected="false">'+e+"</a></li>";n.default(me).append(unescape(escape(r)));var d='<div class="tab-pane fade" id="'+l+'" role="tabpanel" a...

   |
   | It works like this: a POM may declare a repository to use in resolving certain artifacts.
   | However, this repository may have problems with heavy traffic at times, so people have mirrored
   | it to several places.
   |
   | That repository definition will have a unique id, so we can create a mirror reference for that