1
votes

Our pipeline, which has been running fine for a few months, threw the following error today. It looks like the worker pool was unable to be torn down after the pipeline finished:

Aug 31, 2015, 4:35:55 AM (95577a18aba47708): Workflow failed. Causes: (df2004d5f7046091): Step teardown_resource_global_gce_worker_pool2: Resource worker_pool_resource failed to shut down

Job_ID: 2015-08-30_11_01_50-8584391373463411570

Is this a known issue?

1
Sorry for the trouble. We'll investigate. Did your VMs get shut down?Frances
They did, but I've no idea how long it took I'm afraid. Is there any way via the developers console to alert/email if a pipeline fails?Graham Polley

1 Answers

3
votes

There was a transient issue causing this failure which we've identified and should be resolved. We will be adding more checks/resilience to mitigate this issue from impacting dataflow jobs from progressing/shutting down in the future.

Thank you for the report, it was very helpful with our investigation.