Question

我正在尝试使用Databricks CLI创建Databricks作业。该作业运行一个python笔记本。我可以使用“ databricks工作区导入...”在Databricks中创建笔记本。我尝试使用以下方法创建作业： databricks作业创建--json文件C：\ Users \ myid \ Documents \ db_cli \ Hello.json

Hello.json文件包含的位置：

{
  "name": "SimpleJob",
  "settings": {
      "notebook_task": {
        "language": "PYTHON",
        "object_type": "NOTEBOOK",
        "notebook_path": "/Users/Appid@blah.com/SimpleNotebook"
      }
    },
  "new_cluster": {
    "custom_tags": [{"key": "component", "value": "aeops"}],
    "node_type_id": "i3.xlarge",
    "spark_version": "5.5.x-scala2.11",
    "aws_attributes": {
      "ebs_volume_type": "GENERAL_PURPOSE_SSD",
      "ebs_volume_count": 3,
      "ebs_volume_size": 100,
      "spot_bid_price_percent": 120,
      "availability": "SPOT_WITH_FALLBACK",
      "first_on_demand": 1,
      "instance_profile_arn": "arn:aws:iam::<actual acct id goes here>:instance-profile/db-ec2-role",
      "num_workers": 8,
      "spark_env_vars": {
        "PYSPARK_PYTHON": "/databricks/python3/bin/python3"
      }
    },
    "autoscale": {
      "min_workers": 1,
      "max_workers": 100
    },
    "email_notifications": {
      "on_start": [],
      "on_success": [],
      "on_failure": ["johnDoe@blah.com"]
    },
    "timeout_seconds": 0,
    "max_retries": 3,
    "max_concurrent_runs": 1
  }
}

当我跑步时： databricks作业创建--json文件C：\ Users \ user1 \ Documents \ db_cli \ Hello.json

它创建了作业，但是作业中的“任务”没有引用Python笔记本/Users/Appid@blah.com/SimpleNotebook

需要对JSON进行哪些更改，以便我可以创建作业并在作业中引用指定的Python笔记本？

databricks作业创建笔记本INVALID_PARAMETER_VALUE”，“消息”：“必须指定作业设置。”

0 个答案: