* 这里我们传递了 Gunicorn 可以导入和使用的类：

         ```Python
         import uvicorn.workers.UvicornWorker
         ```

* `--bind`：这告诉 Gunicorn 要监听的 IP 和端口，使用冒号 (`:`) 分隔 IP 和端口。
     * 如果您直接运行 Uvicorn，则可以使用`--host 0.0.0.0`和`--port 80`，而不是`--bind 0.0.0.0:80`（Gunicorn 选项）。


在输出中，您可以看到它显示了每个进程的 **PID**（进程 ID）（它只是一个数字）。

你可以看到：

* Gunicorn **进程管理器** 以 PID `19499` 开头（在您的情况下，它将是一个不同的数字）。
* 然后它开始`Listening at: http://0.0.0.0:80`。

    byte[] expectedBytes = buf.clone();
    System.arraycopy(testBytes, 0, expectedBytes, 0, testBytes.length);

    HashingInputStream in = new HashingInputStream(hashFunction, buffer);

    int numOfByteRead = in.read(buf, 0, 100);
    assertEquals(4, numOfByteRead);
    for (int i = 0; i < numOfByteRead; i++) {
      assertEquals(testBytes[i], buf[i]);
    }

    verify(hasher).putBytes(expectedBytes, 0, 4);

  }

  @J2ktIncompatible // TODO(b/259213718): Switch J2kt to String.format("%.4g") once that's supported
  public void testToString() {
    stopwatch.start();
    assertEquals("0.000 ns", stopwatch.toString());
    ticker.advance(1);
    assertEquals("1.000 ns", stopwatch.toString());
    ticker.advance(998);
    assertEquals("999.0 ns", stopwatch.toString());
    ticker.advance(1);

    byte[] expectedBytes = buf.clone();
    System.arraycopy(testBytes, 0, expectedBytes, 0, testBytes.length);

    HashingInputStream in = new HashingInputStream(hashFunction, buffer);

    int numOfByteRead = in.read(buf, 0, 100);
    assertEquals(4, numOfByteRead);
    for (int i = 0; i < numOfByteRead; i++) {
      assertEquals(testBytes[i], buf[i]);
    }

    verify(hasher).putBytes(expectedBytes, 0, 4);

        assertEquals("2004-04-01T12:34:00.000Z", FessFunctions.formatDate(date));

        date = FessFunctions.parseDate("2004-04-01");
        assertEquals("2004-04-01T00:00:00.000Z", FessFunctions.formatDate(date));

        date = FessFunctions.parseDate("2004-04-01T12:34:56.123+09:00");
        assertEquals("2004-04-01T03:34:56.123Z", FessFunctions.formatDate(date));

	if err != nil {
		return 0, 0, 0, buf, err
	}
	metaVer, buf, err := msgp.ReadUint8Bytes(buf)
	if err != nil {
		return 0, 0, 0, buf, err
	}
	if hdrVer > xlHeaderVersion {
		return 0, 0, 0, buf, fmt.Errorf("decodeXLHeaders: Unknown xl header version %d", metaVer)
	}
	if metaVer > xlMetaVersion {
		return 0, 0, 0, buf, fmt.Errorf("decodeXLHeaders: Unknown xl meta version %d", metaVer)
	}

        checkVersionParsing("1.2-alpha-1", 1, 2, 0, 0, "alpha-1");
        checkVersionParsing("1.2-alpha-1-20050205.060708-1", 1, 2, 0, 0, "alpha-1-20050205.060708-1");
        checkVersionParsing("RELEASE", 0, 0, 0, 0, "RELEASE");
        checkVersionParsing("2.0-1", 2, 0, 0, 1, null);

        // 0 at the beginning of a number has a special handling
        checkVersionParsing("02", 0, 0, 0, 0, "02");

```mermaid
graph LR

browser("Browser")
proxy["Proxy on http://0.0.0.0:9999/api/v1/app"]
server["Server on http://127.0.0.1:8000/app"]

browser --> proxy
proxy --> server
```

/// tip

The IP `0.0.0.0` is commonly used to mean that the program listens on all the IPs available in that machine/server.

///

RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt

COPY ./app /code/app

CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "80"]

# If running behind a proxy like Nginx or Traefik add --proxy-headers
# CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "80", "--proxy-headers"]
```

</details>

## 什么是容器

name: Upload SIG Build docker containers regularly

on:
  workflow_dispatch:
  schedule:
      # Run once a week on Sunday at midnight. See http://crontab.guru
      - cron: '0 0 * * 0'
  push:
    paths:
      - '.github/workflows/sigbuild-docker.yml'
      - 'tensorflow/tools/tf_sig_build_dockerfiles/**'
      - '!tensorflow/tools/tf_sig_build_dockerfiles/README.md'
    branches:
      - master