It would be nice if https://cloud.google.com/run/docs/configuring/static-outbound-ip|this documentation mentions the data path taken for the egress traffic for a Cloud Run service instead of me blindly configuring the setup in fact that is what happened and it didn’t work. Seems like the secondary IP of the subnet is used but not sure where (--nat-all-subnet-ip-ranges and --nat-custom-subnet-ip-ranges do the same thing i.e they add both the primary and secondary IP subnet ranges but the latter option which the documentation mentioned did not work it added only primary IP).
Does this page provide sufficient guidance: https://cloud.google.com/nat/docs/set-up-manage-network-address-translation#gcloud|https://cloud.google.com/nat/docs/set-up-manage-network-address-translation#gcloud
Could a comment on the cloud run docs pointing to this for more details be enough?
This captures only the NAT portion. I was talking about the end to end data flow starting with the serverless connector which is GCE VMs behind the scenes and the secondary ip range which seems to be used in this case. Is the secondary IP range associated with the VM?
