我們將於 2025 年 12 月 31 日停止支援 Cloud Deployment Manager。
如果您目前使用 Deployment Manager，請在
2025 年 12 月 31 日前遷移至 Infrastructure Manager 或其他部署技術，確保服務不會中斷。

如要進一步瞭解淘汰和終止運作事宜，請參閱「Deployment Manager 淘汰」。

本頁面由 Cloud Translation API 翻譯而成。

建立容器最佳化的部署作業

本完整範例說明如何透過使用容器最佳化映像檔的虛擬機器建立部署作業。如要進一步瞭解如何使用容器搭配 Compute Engine，請參閱容器最佳化的 Compute Engine 映像檔。

這份逐步導覽會說明如何：

建立簡易的容器資訊清單。
建立使用容器映像檔的設定和範本。
部署資源並且確認已成功完成部署作業。

建立容器資訊清單

如要使用容器，您必須定義容器資訊清單。資訊清單會說明容器映像檔、要啟動的容器、開機時要執行的指令，以及要啟用的通訊埠等屬性。

建立名為 container_manifest.yaml 的檔案，並在當中加入下列內容：

# Copyright 2016 Google Inc. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# This is a container manifest, as described here:
#   https://cloud.google.com/compute/docs/containers/container_vms
apiVersion: v1
kind: Pod
metadata:
  name: simple-echo
spec:
  containers:
    - name: simple-echo
      image: gcr.io/google-samples/hello-app:2.0
      imagePullPolicy: Always
      ports:
        - containerPort: 8080
          hostPort: 8080

這份資訊清單會建立名為 simple-echo 的容器。這個容器會使用 Hello 應用程式容器映像檔，並啟動 Echo 伺服器來監聽通訊埠 8080。

建立範本和設定

接下來，建立透過容器最佳化映像檔啟動虛擬機器執行個體的範本。建立名為 container_vm.[jinja|py] 的檔案，並在當中加入下列內容：

Jinja



{% set COMPUTE_URL_BASE = 'https://www.googleapis.com/compute/v1/' %}
{% set BASE_NAME = env['deployment'] + '-' + env['name'] %}

{% macro GlobalComputeUrl(project, collection, name) -%}
{{ COMPUTE_URL_BASE }}projects/{{ project }}/global/{{ collection }}/{{ name }}
{%- endmacro %}

{% macro ZonalComputeUrl(project, zone, collection, name) -%}
{{ COMPUTE_URL_BASE }}projects/{{ project }}/zones/{{ zone }}/{{ collection }}/{{ name }}
{%- endmacro %}

resources:
- name: {{ BASE_NAME }}
  type: compute.v1.instance
  properties:
    zone: {{ properties['zone'] }}
    machineType: {{ ZonalComputeUrl(env['project'], properties['zone'], 'machineTypes', 'n1-standard-1') }}
    metadata:
      items:
        - key: gce-container-declaration
          value: |
            {{ imports[properties['containerManifest']]|indent(12) }}
    disks:
      - deviceName: boot
        type: PERSISTENT
        autoDelete: true
        boot: true
        initializeParams:
          diskName: {{ BASE_NAME }}-disk
          sourceImage: {{ GlobalComputeUrl('cos-cloud', 'images', properties['containerImage']) }}
    networkInterfaces:
      - accessConfigs:
          - name: external-nat
            type: ONE_TO_ONE_NAT
        network: {{ GlobalComputeUrl(env['project'],  'networks', 'default') }}
    serviceAccounts:
      - email: default
        scopes:
        - https://www.googleapis.com/auth/logging.write
        - https://www.googleapis.com/auth/monitoring.write

Python

# Copyright 2016 Google Inc. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

"""Creates a Container VM with the provided Container manifest."""


COMPUTE_URL_BASE = 'https://www.googleapis.com/compute/v1/'


def GlobalComputeUrl(project, collection, name):
  return ''.join([COMPUTE_URL_BASE, 'projects/', project,
                  '/global/', collection, '/', name])


def ZonalComputeUrl(project, zone, collection, name):
  return ''.join([COMPUTE_URL_BASE, 'projects/', project,
                  '/zones/', zone, '/', collection, '/', name])


def GenerateConfig(context):
  """Generate configuration."""

  res = []
  base_name = (context.env['deployment'] + '-' +
               context.env['name'])

  # Properties for the container-based instance.
  instance = {
      'zone': context.properties['zone'],
      'machineType': ZonalComputeUrl(context.env['project'],
                                     context.properties['zone'],
                                     'machineTypes',
                                     'n1-standard-1'),
      'metadata': {
          'items': [{
              'key': 'gce-container-declaration',
              'value': context.imports[
                  context.properties['containerManifest']],
              }]
      },
      'disks': [{
          'deviceName': 'boot',
          'type': 'PERSISTENT',
          'autoDelete': True,
          'boot': True,
          'initializeParams': {
              'diskName': base_name + '-disk',
              'sourceImage': GlobalComputeUrl('cos-cloud',
                                              'images',
                                              context.properties[
                                                  'containerImage'])
              },
      }],
      'networkInterfaces': [{
          'accessConfigs': [{
              'name': 'external-nat',
              'type': 'ONE_TO_ONE_NAT'
              }],
          'network': GlobalComputeUrl(context.env['project'],
                                      'networks',
                                      'default')
      }],
      'serviceAccounts': [{
          'email': 'default',
          'scopes': [
            "https://www.googleapis.com/auth/logging.write",
            "https://www.googleapis.com/auth/monitoring.write"
          ]
      }]
  }
  res.append({
      'name': base_name,
      'type': 'compute.v1.instance',
      'properties': instance
  })
  # Resources to return.
  resources = {
      'resources': res,
  }

  return resources

建立相對應的結構定義檔案，藉此強制執行範本的結構：

Jinja

# Copyright 2016 Google Inc. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

info:
  title: Container VM
  author: Google Inc.
  description: Creates a Container VM with the provided Container manifest.

required:
  - zone
  - containerImage
  - containerManifest

properties:
  zone:
    description: Zone in which this VM will run
    type: string
  containerImage:
    description: Name of the Google Cloud Container VM Image
    type: string
  containerManifest:
    description: String containing the Container Manifest in YAML
    type: string

Python

# Copyright 2016 Google Inc. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

info:
  title: Container VM
  author: Google Inc.
  description: Creates a Container VM with the provided Container manifest.

required:
  - zone
  - containerImage
  - containerManifest

properties:
  zone:
    description: Zone in which this VM will run
    type: string
  containerImage:
    description: Name of the Google Cloud Container VM Image
    type: string
  containerManifest:
    description: String containing the Container Manifest in YAML
    type: string

請注意，此範本中定義了許多參數，包括：

deployment、name 和 project 環境變數。Deployment Manager 會自動填入這些變數，您無須執行任何操作。
將在設定中定義的 zone、containerImage 和 containerManifest 屬性。

建立一個名為 container_vm.yaml 的設定檔，如下所示：

Jinja

# Copyright 2016 Google Inc. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

imports:
- path: ../../common/container_manifest.yaml
  name: container_manifest
- path: container_vm.jinja

resources:
  - name: my-container-vm
    type: container_vm.jinja
    properties:
      zone: ZONE_TO_RUN
      containerImage: family/cos-stable
      containerManifest: container_manifest

Python

# Copyright 2016 Google Inc. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

imports:
- path: ../../common/container_manifest.yaml
  name: container_manifest
- path: container_vm.py

resources:
  - name: my-container-vm
    type: container_vm.py
    properties:
      zone: ZONE_TO_RUN
      containerImage: family/cos-stable
      containerManifest: container_manifest

請務必將 ZONE_TO_RUN 替換為虛擬機器所需的區域。請注意，這個檔案也定義了要使用的容器映像檔，以及您先前定義的容器資訊清單。

部署虛擬機器執行個體

最後，請使用 Google Cloud CLI 部署虛擬機器執行個體：

gcloud deployment-manager deployments create my-container-deployment \
  --config container_vm.yaml

建立部署作業後，即可檢視其詳細資訊。例如：

$ gcloud deployment-manager deployments describe my-container-deployment
creationTimestamp: '2015-04-02T12:24:31.645-07:00'
fingerprint: ''
id: '8602410885927938432'
manifest: https://www.googleapis.com/deploymentmanager/v2/projects/myproject/global/deployments/my-container-deployment/manifests/manifest-1428002671645
name: my-container-deployment
state: DEPLOYED
resources:
NAME                                     TYPE                 ID                   UPDATE_STATE  ERRORS
my-container-deployment-my-container-vm  compute.v1.instance  3775082612632070557  COMPLETED     -

確認您的執行個體正在運作

如要測試容器執行個體是否已經啟動，請在瀏覽器中前往虛擬機器的外部 IP 位址，畫面上應該會顯示 hello world：

新增 Compute Engine 防火牆規則，藉此允許您查詢虛擬機器中通過通訊埠 8080 的流量：
```
gcloud compute firewall-rules create allow-8080 --allow tcp:8080
```

取得執行個體的外部 IP 位址：

$ gcloud compute instances describe my-container-deployment-my-container-vm
...
name: my-container-vm-my-container-deployment
networkInterfaces:
- accessConfigs:
  - kind: compute#accessConfig
    name: external-nat
    natIP: 104.197.8.138
    type: ONE_TO_ONE_NAT
  name: nic0
  network: https://www.googleapis.com/compute/v1/projects/myproject/global/networks/default
  networkIP: 10.240.97.220
scheduling:
  automaticRestart: true
  onHostMaintenance: MIGRATE
selfLink: https://www.googleapis.com/compute/v1/projects/myproject/zones/us-central1-a/instances/my-container-deployment-my-container-vm
status: RUNNING
tags:
  fingerprint: 42WmSpB8rSM=
zone: https://www.googleapis.com/compute/v1/projects/myproject/zones/us-central1-a
...

在本範例中，外部 IP 為 104.197.8.138。

在瀏覽器視窗的瀏覽器列中輸入外部 IP 和通訊埠 8080。例如 104.197.8.138:8080。

如果成功，您應該會看到 hello world 訊息。

(選用) 刪除部署作業

如果要節省成本而且已不需要部署作業，您可將其刪除。

gcloud deployment-manager deployments delete my-container-deployment

後續步驟

如要進一步瞭解 Deployment Manager，請參閱完整的使用手冊或透過 API 瞭解詳情。

試試看一些其他的教學課程：

建立容器最佳化的部署作業 透過集合功能整理內容 你可以依據偏好儲存及分類內容。

建立容器資訊清單

建立範本和設定

Jinja

Python

Jinja

Python

Jinja

Python

部署虛擬機器執行個體

確認您的執行個體正在運作

(選用) 刪除部署作業

後續步驟

建立容器最佳化的部署作業