在作为 Google Cloud Function 运行的 Python 脚本中使用 gcloud 命令时出错

Question

I have below python code running in Google Cloud function.

pname = format(os.environ.get('GCP_PROJECT'))
project_metadata_file = "/tmp/" + pname + "_proj_metadata.json"
bucket_name = "XXXX"
H = open(project_metadata_file, "w")
project_list_command = "gcloud projects describe " + pname + " --format json"
project_output = subprocess.check_output(shlex.split(project_list_command))
project_output_json = json.loads(project_output)
H.write(project_output_json)
H.close()
upload_blob(bucket_name, project_metadata_file, os.path.basename(project_metadata_file))
return pname

Here is the error I am getting from the Cloud function. Could you please advise how can I have gcloud as a requirement for this cloud function. Please advise.

Error: function terminated. Recommended action: inspect logs for termination reason. Details:
[Errno 2] No such file or directory: 'gcloud': 'gcloud'

Thanks

Answer 1

You have to expect your function as being deployed into a container that includes, in your case, a Python runtime. You can not assume that your cloud function includes a shell environment with cloud sdk installed. I would recommend using Method: projects.get API . In case you want to use the cloud sdk commands then Google Cloud Run would be one of the solutions.

Enter the project name and you will receive this output:

{
  "projectNumber": "",
  "projectId": "",
  "lifecycleState": "",
  "name": "",
  "createTime": "",
  "parent": {
    "type": "",
    "id": ""
  }
}

Answer 2

As DazWilkin already mentioned, your code will be much cleaner and robust if you use specialized library.

If you really need to use Cloud SDK with your function, you can achieve it with the following.

1. Download the appropriate archive .
2. Extract it to your function's folder
3. Set path to gcloud command to './google-cloud-sdk/bin/gcloud' in the code. This may vary depending on where you extracted SDK in your function folder.
4. Deploy function with either gcloud command or zip it and deploy from Console.