private final List<Header> requestHeaderList = new ArrayList<>();

    /** Map of HTTP client properties */
    private final Map<String, Object> httpClientPropertyMap = new HashMap<>();

    /** Task for monitoring idle connections */
    private TimeoutTask connectionMonitorTask;

    /** Connection timeout in milliseconds */
    protected Integer connectionTimeout;

    opt Tasks
        operation -->> tasks: Send background tasks
    end
    opt Raise other exception
        tasks -->> tasks: Handle exceptions in the background task code
    end
```

/// info | 说明

只会向客户端发送 **一次响应** ，可能是一个错误响应，也可能是来自 *路由函数* 的响应。

在发送了其中一个响应之后，就无法再发送其他响应了。

///

/// tip | 提示

 *  Fix: Handle multiple 1xx responses.

 *  Fix: Address a performance bug in our internal task runner. We had a race condition that could
    result in it OkHttp starting a thread for each queued task, even when a single thread could run
    all of them.

 *  Fix: Address a performance bug in `MultipartReader`. We were scanning the entire input stream

        thresholdAdjustmentTask = scheduler.scheduleWithFixedDelay(this::performDynamicThresholdAdjustment, 30, 30, TimeUnit.SECONDS);
        log.debug("[{}] Started dynamic threshold adjustment task", name);
    }

    private void performDynamicThresholdAdjustment() {
        try {
            RollingWindow.WindowMetrics metrics = rollingWindow.getMetrics();

            if (logger.isInfoEnabled()) {
                logger.info("Finished Crawler.");
            }

            return Constants.EXIT_OK;
        } catch (final Throwable t) {
            logger.warn("Crawl task failed with an exception.", t);
            return Constants.EXIT_FAIL;
        } finally {
            pathMappingHelper.removePathMappingList(options.sessionId);

 * small portion of these would be in use.
 *
 * <p>Prior to this class, one might be tempted to use {@code Map<K, Lock>}, where {@code K}
 * represents the task. This maximizes concurrency by having each unique key mapped to a unique
 * lock, but also maximizes memory footprint. On the other extreme, one could use a single lock for

	Err  error     // Error returned by the remote peer for an rpc call
}

// A NotificationGroup is a collection of goroutines working on subtasks that are part of
// the same overall task.
//
// A zero NotificationGroup is valid and does not cancel on error.
type NotificationGroup struct {
	workers    *workers.Workers
	errs       []NotificationPeerErr
	retryCount int
}

        }
        TimeoutManager.getInstance().addTimeoutTarget(() -> {
            if (logger.isDebugEnabled()) {
                logger.debug("[scheduleParentGroupLookup] Async task started for user: {}", user.getName());
            }
            final long startTime = System.currentTimeMillis();
            try {
                final List<String> updatedGroups = new ArrayList<>(initialGroups);

    opt Tasks
        operation -->> tasks: Send background tasks
    end
    opt Raise other exception
        tasks -->> tasks: Handle exceptions in the background task code
    end
```

/// info | 정보

클라이언트에 **하나의 응답** 만 전송됩니다. 이는 오류 응답 중 하나일 수도 있고,*경로 작업*에서 생성된 응답일 수도 있습니다.

이러한 응답 중 하나가 전송된 후에는 다른 응답을 보낼 수 없습니다.

///

/// tip | 팁

Ancak FastAPI olmadan `async` / `await` kullanmak istiyorsanız, <a href="https://docs.python.org/3/library/asyncio-task.html#coroutine" class="external-link" target="_blank">resmi Python belgelerini kontrol edin</a>.

### Asenkron kodun diğer biçimleri

Bu `async` ve `await` kullanimi oldukça yenidir.