3
votes

I was orchestrating two dataflow job with cloud composer and it was working fine for month. Suddenly the two jobs stopped working with the following error message:

in download_blob File "/usr/local/lib/python3.6/site-packages/google/cloud/storage/client.py", line 399, in get_bucket retry=retry, File "/usr/local/lib/python3.6/site-packages/google/cloud/storage/bucket.py", line 1002, in reload retry=retry, File "/usr/local/lib/python3.6/site-packages/google/cloud/storage/_helpers.py", line 225, in reload retry=retry, File "/usr/local/lib/python3.6/site-packages/google/cloud/storage/_http.py", line 63, in api_request return call() File "/usr/local/lib/python3.6/site-packages/google/api_core/retry.py", line 286, in retry_wrapped_func on_error=on_error, File "/usr/local/lib/python3.6/site-packages/google/api_core/retry.py", line 184, in retry_target return target() File "/usr/local/lib/python3.6/site-packages/google/cloud/_http.py", line 479, in api_request timeout=timeout, File "/usr/local/lib/python3.6/site-packages/google/cloud/_http.py", line 337, in _make_request method, url, headers, data, target_object, timeout=timeout File "/usr/local/lib/python3.6/site-packages/google/cloud/_http.py", line 374, in _do_request return self.http.request( File "/usr/local/lib/python3.6/site-packages/google/cloud/_http.py", line 157, in http return self._client._http File "/usr/local/lib/python3.6/site-packages/google/cloud/client.py", line 187, in _http self._http_internal.configure_mtls_channel(self._client_cert_source) AttributeError: 'AuthorizedSession' object has no attribute 'configure_mtls_channel'

In the jobs I download a file from google cloud storage with the storage client. I assumed it was because of some dependencies issues. In the composer environment I installed google-cloud-storage without specifying a version. I tried specifying different versions of the package but nothing seems to work.

Thanks!

1
Can you provide more details on how you were able to debug this issue? so I can update my answer accordingly.Ismail
I pinned google-cloud-core to 1.5.0 as you suggest and it solved the problem. And it was a batch job. Thanks!ShaqL

1 Answers

2
votes

This seems to be related to this issue.

Try pinning google-cloud-core to 1.5.0, then I highly recommend for you to Drain your jobs once you get them back to work (assuming they have streaming jobs).