You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
All services start successfully and the MLFlow UI can be seen on http://127.0.0.1:5001/ and Jupyter notebooks on port 8888. However, when I attempt to run the included notebook mlflow_example.ipynb I get the error below. Worth noting that the metrics and parameters log ok and are visible but the artifact store is not working ... any help appreciated.
---------------------------------------------------------------------------
PermissionError Traceback (most recent call last)
Cell In [1], line 12
10 for i in range(5):
11 mlflow.log_metric("i",i,step=i)
---> 12 mlflow.log_artifact("mlflow_example.ipynb")
File /opt/conda/lib/python3.9/site-packages/mlflow/tracking/fluent.py:869, in log_artifact(local_path, artifact_path)
846 """
847 Log a local file or directory as an artifact of the currently active run. If no run is
848 active, this method will create a new active run.
(...)
866 mlflow.log_artifact("features.txt")
867 """
868 run_id = _get_or_start_run().info.run_id
--> 869 MlflowClient().log_artifact(run_id, local_path, artifact_path)
File /opt/conda/lib/python3.9/site-packages/mlflow/tracking/client.py:1091, in MlflowClient.log_artifact(self, run_id, local_path, artifact_path)
1056 def log_artifact(self, run_id, local_path, artifact_path=None) -> None:
1057 """
1058 Write a local file or directory to the remote ``artifact_uri``.
1059
(...)
1089 is_dir: False
1090 """
-> 1091 self._tracking_client.log_artifact(run_id, local_path, artifact_path)
File /opt/conda/lib/python3.9/site-packages/mlflow/tracking/_tracking_service/client.py:456, in TrackingServiceClient.log_artifact(self, run_id, local_path, artifact_path)
454 artifact_repo.log_artifacts(local_path, path_name)
455 else:
--> 456 artifact_repo.log_artifact(local_path, artifact_path)
File /opt/conda/lib/python3.9/site-packages/mlflow/store/artifact/sftp_artifact_repo.py:90, in SFTPArtifactRepository.log_artifact(self, local_file, artifact_path)
88 artifact_dir = posixpath.join(self.path, artifact_path) if artifact_path else self.path
89 with self.pool.get_sfp_connection() as sftp:
---> 90 sftp.makedirs(artifact_dir)
91 sftp.put(local_file, posixpath.join(artifact_dir, os.path.basename(local_file)))
File /opt/conda/lib/python3.9/site-packages/pysftp/__init__.py:699, in Connection.makedirs(self, remotedir, mode)
697 head, tail = os.path.split(remotedir)
698 if head and not self.isdir(head):
--> 699 self.makedirs(head, mode)
701 if tail:
702 self.mkdir(remotedir, mode=mode)
File /opt/conda/lib/python3.9/site-packages/pysftp/__init__.py:699, in Connection.makedirs(self, remotedir, mode)
697 head, tail = os.path.split(remotedir)
698 if head and not self.isdir(head):
--> 699 self.makedirs(head, mode)
701 if tail:
702 self.mkdir(remotedir, mode=mode)
File /opt/conda/lib/python3.9/site-packages/pysftp/__init__.py:702, in Connection.makedirs(self, remotedir, mode)
699 self.makedirs(head, mode)
701 if tail:
--> 702 self.mkdir(remotedir, mode=mode)
File /opt/conda/lib/python3.9/site-packages/pysftp/__init__.py:626, in Connection.mkdir(self, remotepath, mode)
614 """Create a directory named remotepath with mode. On some systems,
615 mode is ignored. Where it is used, the current umask value is first
616 masked out.
(...)
623
624 """
625 self._sftp_connect()
--> 626 self._sftp.mkdir(remotepath, mode=int(str(mode), 8))
File /opt/conda/lib/python3.9/site-packages/paramiko/sftp_client.py:460, in SFTPClient.mkdir(self, path, mode)
458 attr = SFTPAttributes()
459 attr.st_mode = mode
--> 460 self._request(CMD_MKDIR, path, attr)
File /opt/conda/lib/python3.9/site-packages/paramiko/sftp_client.py:857, in SFTPClient._request(self, t, *args)
855 def _request(self, t, *args):
856 num = self._async_request(type(None), t, *args)
--> 857 return self._read_response(num)
File /opt/conda/lib/python3.9/site-packages/paramiko/sftp_client.py:909, in SFTPClient._read_response(self, waitfor)
906 if num == waitfor:
907 # synchronous
908 if t == CMD_STATUS:
--> 909 self._convert_status(msg)
910 return t, msg
912 # can not rewrite this to deal with E721, either as a None check
913 # nor as not an instance of None or NoneType
File /opt/conda/lib/python3.9/site-packages/paramiko/sftp_client.py:940, in SFTPClient._convert_status(self, msg)
938 raise IOError(errno.ENOENT, text)
939 elif code == SFTP_PERMISSION_DENIED:
--> 940 raise IOError(errno.EACCES, text)
941 else:
942 raise IOError(text)
PermissionError: [Errno 13] Permission denied
The text was updated successfully, but these errors were encountered:
All services start successfully and the MLFlow UI can be seen on http://127.0.0.1:5001/ and Jupyter notebooks on port 8888. However, when I attempt to run the included notebook
mlflow_example.ipynb
I get the error below. Worth noting that the metrics and parameters log ok and are visible but the artifact store is not working ... any help appreciated.The text was updated successfully, but these errors were encountered: