Open amygdala opened 5 years ago
I think https://github.com/kubeflow/examples/pull/621 should fix it.
It is likely to be the istio side car hasn't started
https://github.com/kubeflow/examples/pull/621/files#diff-19caa32109d22abdba8778f600e00f72R342
I also saw this error when the cluster had been sitting idle for a few days. This seems like a very common error. Would it make sense to have the client itself catch the error and retry, rather than requiring error mgmt in the user code?
Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/urllib3/connectionpool.py", line 603, in urlopen
chunked=chunked)
File "/usr/local/lib/python3.6/dist-packages/urllib3/connectionpool.py", line 387, in _make_request
six.raise_from(e, None)
File "<string>", line 2, in raise_from
File "/usr/local/lib/python3.6/dist-packages/urllib3/connectionpool.py", line 383, in _make_request
httplib_response = conn.getresponse()
File "/usr/lib/python3.6/http/client.py", line 1331, in getresponse
response.begin()
File "/usr/lib/python3.6/http/client.py", line 297, in begin
version, status, reason = self._read_status()
File "/usr/lib/python3.6/http/client.py", line 266, in _read_status
raise RemoteDisconnected("Remote end closed connection without"
http.client.RemoteDisconnected: Remote end closed connection without response
/area engprod /priority p1
/kind bug
KF 0.6.1, GKE, using IAP
Running the example notebook , I've seen the following error each time.
Re-running the cell fixes things, so maybe a retry is needed.
With this code:
I initially get this error (again, a rerun of the cell lets it go through):
Here's the full trace: https://gist.github.com/amygdala/19670fcf32c1369c03d4125e86db822b