Has anyone ran into this error when running a Flow Prefect v

Question

Has anyone ran into this error when running a Flow Prefect version 0 12 1 on a Kubernetes Agent ` 2020 07 01 13 40 32 DEBUG prefect CloudFlowRunner | Failed to retrieve flow state with error AttributeError SSLSocket object has no attribute connection ` The Flow runs fine on my local machine and the healthcheck passes but when it s deployed to K8 it seems the CloudFlowRunner cannot write logs to Prefect Cloud and aborts the execution of the Flow

josh · Answer

thinking face I have not are you using core s server or Cloud

Scott Zelenka · Answer

Prefect Cloud

Scott Zelenka · Answer

that error shows up on the Flow Run s container after it pulls the Docker image and attempts to start the Flow Run

Scott Zelenka · Answer

The full logs are ``` 2020 07 01 13 40 32 DEBUG prefect CloudFlowRunner | Failed to retrieve flow state with error AttributeError SSLSocket object has no attribute connection 2020 07 01 13 40 32 CRITICAL CloudHandler | Failed to write log with error SSLSocket object has no attribute connection 2020 07 01 13 40 32 CRITICAL CloudHandler | Failed to write log with error SSLSocket object has no attribute connection 2020 07 01 13 40 32 CRITICAL CloudHandler | Failed to write log with error SSLSocket object has no attribute connection 2020 07 01 13 40 32 CRITICAL CloudHandler | Failed to write log with error SSLSocket object has no attribute connection 2020 07 01 13 40 32 CRITICAL CloudHandler | Unable to write logs to Prefect Cloud 2020 07 01 13 40 32 CRITICAL CloudHandler | Unable to write logs to Prefect Cloud 2020 07 01 13 40 32 CRITICAL CloudHandler | Unable to write logs to Prefect Cloud 2020 07 01 13 40 32 CRITICAL CloudHandler | Unable to write logs to Prefect Cloud```

Scott Zelenka · Answer

I m thinking it s a version dependency issue the Python packages for that Flow Run are ``` ujson==3 0 0 requests==2 24 0 pandas==1 0 5 numpy==1 19 0 sqlalchemy==1 3 18 hvac==0 10 4 snowflake connector python==2 2 8 snowflake sqlalchemy==1 2 3 pyarrow==0 17 1 ```

josh · Answer

I m thinking it might be something urllib3 version related

josh · Answer

Yeah ^

josh · Answer

Any way you can see which version of urllib3 is being installed in that image I m seeing it being related to the snowflake connector and a new version was just released a week ago

Scott Zelenka · Answer

I ll download it locally

Scott Zelenka · Answer

urllib3==1 25 9

josh · Answer

Does that differ compared to the one you have installed locally

Scott Zelenka · Answer

nope I m creating a fresh venv to test again on my local machine

Scott Zelenka · Answer

so the versions are the same but the difference being when I run locally I m just doing the `flow run ` while the other is using the CloudRunner

josh · Answer

Ah yeah so that makes sense as to why it wouldn t happen locally since it s not actually calling out to cloud Two things 1 You could try doing a cloud run in that venv using local storage + local agent to check your dependencies 2 Open an issue with your dependencies and we can try to recreate

josh · Answer

You could also try pinning the snowflake connector ` lt 2 2 8` and see if that has any effect

Scott Zelenka · Answer

rolling it back to 2 2 7 fixed that issue I ll log something in GitHub later today to hopefully catch these types of dependency errors during the healthcheck

josh · Answer

Great < Scott Zelenka>