Reputation: 13078
I'm suddenly unable to run Tensorboard on my Google Cloud Platform account.
my-username@my-projectp:~$ tensorboard --logdir gs://my_bucket --host localhost --port 8088
TensorBoard 0.4.0rc3 at http://localhost:8088 (Press CTRL+C to quit)
Exception in thread Reloader:
Traceback (most recent call last):
File "/usr/lib/python2.7/threading.py", line 810, in __bootstrap_inner
self.run()
File "/usr/lib/python2.7/threading.py", line 763, in run
self.__target(*self.__args, **self.__kwargs)
File "/usr/local/lib/python2.7/dist-packages/tensorboard/backend/application.py", line 350, in _reload_forever
reload_multiplexer(multiplexer, path_to_run)
File "/usr/local/lib/python2.7/dist-packages/tensorboard/backend/application.py", line 322, in reload_multiplexer
multiplexer.AddRunsFromDirectory(path, name)
File "/usr/local/lib/python2.7/dist-packages/tensorboard/backend/event_processing/plugin_event_multiplexer.py", line 175, in AddRunsFromDirectory
for subdir in GetLogdirSubdirectories(path):
File "/usr/local/lib/python2.7/dist-packages/tensorboard/backend/event_processing/plugin_event_multiplexer.py", line 445, in <genexpr>
subdir
File "/usr/local/lib/python2.7/dist-packages/tensorboard/backend/event_processing/io_wrapper.py", line 50, in ListRecursively
for dir_path, _, filenames in tf.gfile.Walk(top):
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/lib/io/file_io.py", line 518, in walk
for subitem in walk(os.path.join(top, subdir), in_order):
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/lib/io/file_io.py", line 518, in walk
for subitem in walk(os.path.join(top, subdir), in_order):
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/lib/io/file_io.py", line 518, in walk
for subitem in walk(os.path.join(top, subdir), in_order):
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/lib/io/file_io.py", line 518, in walk
for subitem in walk(os.path.join(top, subdir), in_order):
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/lib/io/file_io.py", line 499, in walk
listing = list_directory(top)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/lib/io/file_io.py", line 478, in list_directory
compat.as_bytes(dirname), status)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/errors_impl.py", line 473, in __exit__
c_api.TF_GetCode(self.status.status))
PermissionDeniedError: /etc/docker; Permission denied
How can this be fixed?
Upvotes: 0
Views: 1038
Reputation: 13078
Found the answer. Looks like a bug in Cloud Shell or Tensorflow. The problem was that I added gs://
twice in my Tensorflow's model-dir
value. So some of the logs that Tensorflow tried to read were under a path like
gs://my_bucket/models/model_xyz/gs://my_bucket/models/model_xyz
There was indeed a directory called gs:
in my Cloud Storage account. Inside of which there was another directory named /
. Deleting the gs:
directory fixed the issue.
Upvotes: 1