return failFastResponse!!
    }

    val result = responseQueue.take()

    // If take() returned because we're shutting down, then enqueue another dead letter so that any
    // other threads waiting on take() will also return.
    if (result == DEAD_LETTER) responseQueue.add(DEAD_LETTER)

    return result
  }

  override fun peek(): MockResponse {
    return responseQueue.peek() ?: failFastResponse ?: super.peek()
  }

And if the server is restarted (for example after updates, or migrations from the cloud provider) you probably **won't notice it**. And because of that, you won't even know that you have to restart the process manually. So, your API will just stay dead. 😱

### Run Automatically on Startup

* Then it detects that it has to use the worker class at `uvicorn.workers.UvicornWorker`.
* And then it starts **4 workers**, each with its own PID: `19511`, `19513`, `19514`, and `19515`.

Gunicorn would also take care of managing **dead processes** and **restarting** new ones if needed to keep the number of workers. So that helps in part with the **restart** concept from the list above.

    val response = call.execute()
    val inputStream = response.body.byteStream()
    assertThat(inputStream.read()).isEqualTo('A'.code)
    assertThat(inputStream.read()).isEqualTo('B'.code)
    assertThat(inputStream.read()).isEqualTo('C'.code)
    assertThat(inputStream.read()).isEqualTo(-1)
    assertThat(inputStream.read()).isEqualTo(-1)
    inputStream.close()
  }

  @ParameterizedTest

    /**
     * Attaches [tag] to the request using [T] as a key. Tags can be read from a request using
     * [Request.tag]. Use null to remove any existing tag assigned for [T].
     *
     * Use this API to attach timing, debugging, or other application data to a request so that
     * you may read it in interceptors, event listeners, or callbacks.
     */
    @JvmName("reifiedTag")

- Names inside ZIP files are kept unmodified, but some may lead to invalid paths. See [Object key naming guidelines](https://docs.aws.amazon.com/AmazonS3/latest/userguide/object-keys.html) on safe names.
- This API behavior is limited for following **read** operations on files inside a zip archive:
  - `HeadObject`
  - `GetObject`
  - `ListObjectsV2`

affinity towards odd number erasure sets to provide for uniform distribution across nodes*. This is to ensure that same number of drives are pariticipating in any erasure set. For example if you have 2 nodes with 180 drives then GCD is 15 but this would lead to uneven distribution, one of the nodes would participate more drives. To avoid this the affinity is given towards nodes which leads to next best GCD factor of 12 which provides uniform distribution.

- In this algorithm, we also make sure...

    val buffer = Buffer()
    operator.read(6, buffer, 21)
    operator.read(36, buffer, 1)
    operator.read(5, buffer, 5)
    operator.read(28, buffer, 8)
    operator.read(17, buffer, 10)
    operator.read(36, buffer, 2)
    operator.read(2, buffer, 4)
    operator.write(0, buffer, buffer.size)
    operator.read(0, buffer, 12)
    operator.read(47, buffer, 3)
    operator.read(45, buffer, 2)
    operator.read(47, buffer, 3)

  - release-managers
  - AuraSinis # 1.24 Release Notes Lead
  - cici37 # 1.23 Release Notes Lead
  - csantanapr # 1.25 Release Notes Lead
  - harshanarayana # 1.27 Release Notes Lead
  - ramrodo # 1.26 Release Notes Lead
  - sanchita-07 # 1.28 Release Notes Lead
  - fsmunoz # 1.29 Release Notes Lead
  - rashansmith # 1.30 Release Notes Lead
reviewers:
  - release-managers
  - fykaa # 1.30 Release Notes Shadow

Requests come from the client specifically asking for a set of resources. This could be requesting the initial set of resources on a new connection, or from a new dependency. For example, a push of `Cluster X` referencing `Endpoint Y` may lead to a request for `Endpoint Y` if it is not already known to the client.