Analytic Fails to call SklearnPipe_train


I have successfully trained my pipeline manually through the console. I was hoping to set something up within an analytic to train or retrain. However, I get the following error. It looks as though the analytic container engine has worse permissions than I do.

"unable to call processSource for analytic CPIAlertsGBMAnalytic source EngineeringItem.EDS|1234567:

C3RuntimeException: Error executing command: /usr/local/share/c3/condaEnvs/CatDataLake/catkinska/py-pipeline2/bin/python /tmp/
Traceback (most recent call last):
  File "/tmp/", line 1465, in <module>
    _c3_result = train(this = _c3_inputs.get('this'),input = _c3_inputs.get('input'),targetOutput = _c3_inputs.get('targetOutput'),spec = _c3_inputs.get('spec'))
  File "/tmp/", line 1321, in train
    if not this.type.isTrainable(this=dict_with_type_at_front(this)):
  File "/tmp/", line 134, in __call__
    return  connection.request_json(self._type_name, self._function_name, kwargs)
  File "/tmp/", line 92, in request_json
    raise e
urllib2.HTTPError: HTTP Error 401: Unauthorized

  from action SklearnPipe.train
  from env_server.js, line 129
    127    if (C3._context.locale)
    128      options.language = C3._context.locale;
  > 129    return c3CallAction(target, args, options);
    130  }


It has been resolved on 7.8


That’s great! I’m on 7.8.41. :smiley:



Currently, you will be on c3server v7.8.2.262 today evening. For this fix, you would need c3server v7.8.2.366.

Let’s discuss.