Longhorn 雲原生容器分散式儲存 - Python Client

為少 發表於 2021-09-03
Python

Longhorn 雲原生容器分散式儲存 - Python Client

內容來源於官方 Longhorn 1.1.2 英文技術手冊。

系列

目前,您可以使用 Longhorn UI 操作 Longhorn。同時,您可以使用 Python 訪問 Longhorn API,如下所示。

  1. 獲取 Longhorn API 端點

    Longhorn 通訊的一種方式是通過 longhorn-frontend service。

    如果您在安裝 Longhorn 的同一叢集中執行自動化/指令碼(automation/scripting)工具,請連線到端點 http://longhorn-frontend.longhorn-system/v1

    如果您在本地機器上執行自動化/指令碼(automation/scripting)工具,請使用 kubectl port-forwardlonghorn-frontend service 轉發到 localhost

    kubectl port-forward services/longhorn-frontend 8080:http -n longhorn-system
    

    並連線到端點 http://localhost:8080/v1

  2. 使用 Python Client

    longhorn.py 檔案(包含 Python client)匯入到以下 Python 指令碼中,並從 API 端點建立一個 client

    https://github.com/longhorn/longhorn-tests/blob/master/manager/integration/tests/longhorn.py

    import longhorn
    
    # If automation/scripting tool is inside the same cluster in which Longhorn is installed
    longhorn_url = 'http://longhorn-frontend.longhorn-system/v1'
    # If forwarding `longhorn-frontend` service to localhost
    longhorn_url = 'http://localhost:8080/v1'
    
    client = longhorn.Client(url=longhorn_url)
    
    # Volume operations
    # List all volumes
    volumes = client.list_volume()
    # Get volume by NAME/ID
    testvol1 = client.by_id_volume(id="testvol1")
    # Attach TESTVOL1
    testvol1 = testvol1.attach(hostId="worker-1")
    # Detach TESTVOL1
    testvol1.detach()
    # Create a snapshot of TESTVOL1 with NAME
    snapshot1 = testvol1.snapshotCreate(name="snapshot1")
    # Create a backup from a snapshot NAME
    testvol1.snapshotBackup(name=snapshot1.name)
    # Update the number of replicas of TESTVOL1
    testvol1.updateReplicaCount(replicaCount=2)
    # Find more examples in Longhorn integration tests https://github.com/longhorn/longhorn-tests/tree/master/manager/integration/tests
    
    # Node operations
    # List all nodes
    nodes = client.list_node()
    # Get node by NAME/ID
    node1 = client.by_id_node(id="worker-1")
    # Disable scheduling for NODE1
    client.update(node1, allowScheduling=False)
    # Enable scheduling for NODE1
    client.update(node1, allowScheduling=True)
    # Find more examples in Longhorn integration tests https://github.com/longhorn/longhorn-tests/tree/master/manager/integration/tests
    
    # Setting operations
    # List all settings
    settings = client.list_setting()
    # Get setting by NAME/ID
    backupTargetsetting = client.by_id_setting(id="backup-target")
    # Update a setting
    backupTargetsetting = client.update(backupTargetsetting, value="s3://backupbucke[email protected]/")
    # Find more examples in Longhorn integration tests https://github.com/longhorn/longhorn-tests/tree/master/manager/integration/tests
    
公眾號:黑客下午茶

相關文章