"Workflow failed. Causes: Network default is not accessible to Dataflow Service account or does not exist"

0
votes

I deleted my default network and working on my own custom network. I imported training data set in Google cloud datalab. When I try to run python code to create ML data set using Dataflow I am getting above error.

Training set I imported is babyweight.

I created instance for datalab in custom network by giving --subnetwork=custom subnet variable

Firewall already open for all ports and all sources on the network.

Service account for Dataflow service agent and cloud ml service agent also have compute network user role.

I also try to set subnetwork variable with complete URL in python code as well as on os.envrion['subnetwork'].

But whenever click on RUN cell I get same error. Can you help me out how to resolve this error. I can create default network again but I am doing this for learning purpose. Can you help me out to resolve the issue

1
cloudgoogle-cloud-dataflowdataflow
In order to investigate further could you share your python code and the full error message? - Alexandre Moraes
Hi, Thanks for reaching. I am able to resolve the issue.I was passing the subnetwork in wrong parameters. After adding subnetwork in pipeline argument issue got resolve. Now I am able to run the job. - Ayush Goel

1 Answers

0
votes

I was adding subnetwork variable in pipeline options in wrong way. Issue got resolve after correcting the parameters.