of unacknowledged data per stream and no per-connection limit.

 *  Fix: Don't operate on a connection after it's been returned to the pool. This race occurred
    on failed web socket connection attempts.

 *  Upgrade: [Okio 3.6.0][okio_3_6_0].

 *  Upgrade: [Kotlin 1.8.21][kotlin_1_8_21].


## Version 4.11.0

_2023-04-22_

	}

	toEncode := io.Reader(data)
	if data.Size() >= bigFileThreshold {
		// We use 2 buffers, so we always have a full buffer of input.
		pool := globalBytePoolCap.Load()
		bufA := pool.Get()
		bufB := pool.Get()
		defer pool.Put(bufA)
		defer pool.Put(bufB)
		ra, err := readahead.NewReaderBuffer(data, [][]byte{bufA[:fi.Erasure.BlockSize], bufB[:fi.Erasure.BlockSize]})
		if err == nil {
			toEncode = ra

});
```

## Performance Tuning

### Thread Configuration

```java
// Optimize thread pool settings
crawler.crawlerContext.setNumOfThread(20);           // Number of crawler threads
crawler.crawlerContext.setMaxThreadCheckCount(50);   // Thread monitoring frequency
```

### Connection Pool Tuning

```java
container.singleton("httpClient", HcHttpClient.class, client -> {

	"bytes"
	"encoding/json"
	"fmt"
	"sync"
	"sync/atomic"
)

// GetByteBuffer returns a byte buffer from the pool.
var GetByteBuffer = func() []byte {
	return make([]byte, 0, 4096)
}

// Sub - subscriber entity.
type Sub[T Maskable] struct {
	ch     chan T
	types  Mask
	filter func(entry T) bool
}

// PubSub holds publishers and subscribers
type PubSub[T Maskable, M Maskable] struct {
	// atomics, keep at top:

          // We have a connected TCP connection. Cancel and defer the racing connects that all lost.
          cancelInFlightConnects()

          // Finish connecting. We won't have to if the winner is from the connection pool.
          if (!connectResult.plan.isReady) {
            connectResult = connectResult.plan.connectTlsEtc()
          }

          if (connectResult.isSuccess) {
            return connectResult.plan.handleSuccess()

 * your HTTP calls. This is because each client holds its own connection pool and thread pools.
 * Reusing connections and threads reduces latency and saves memory. Conversely, creating a client
 * for each request wastes resources on idle pools.
 *
 * Use `new OkHttpClient()` to create a shared instance with the default settings:
 *
 * ```java
 * // The singleton HTTP client.

 * governing permissions and limitations under the License.
 */
package org.codelibs.fess.crawler.pool;

import org.apache.commons.pool2.BasePooledObjectFactory;
import org.apache.commons.pool2.PooledObject;
import org.apache.commons.pool2.impl.DefaultPooledObject;
import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;

   *    can successfully complete the cache entry.
   *
   *  * Throttle the response. The HTTP/1 connection pool will attempt to read the entire response
   *    body so that it can pool the connection.
   */
  private fun trailersWithClientPrematureClose(protocol: Protocol) {
    val halfResponseBody = "a".repeat(OKHTTP_CLIENT_WINDOW_SIZE)
    enableProtocol(protocol)

 *  **There is no longer a global singleton connection pool.** In OkHttp 2.x,
    all `OkHttpClient` instances shared a common connection pool by default.
    In OkHttp 3.x, each new `OkHttpClient` gets its own private connection pool.
    Applications should avoid creating many connection pools as doing so
    prevents connection reuse. Each connection pool holds its own set of

			// No ellipses pattern. Anonymize host name from every pool arg
			pools := strings.Fields(poolsArgs)
			anonPools = make([]string, len(pools))
			for index, arg := range pools {
				anonPools[index] = anonAddr(arg)
			}
			return cmdLineWithoutPools + strings.Join(anonPools, " ")
		}

		// Ellipses pattern in pool args. Regex groups:
		// 1 - server prefix
		// 2 - number sequence for servers