Collab with network-edge to pin-point what they'd like to capture and how to make it happen.

Prior art:

tcpdump service step: https://github.com/openshift/release/tree/master/ci-operator/step-registry/tcpdump-service

Possibly the a similar DaemonSet as the one used by Grant Spence and Hongkai when debugging the CI cluster DNS issues? See https://github.com/gcs278/openshift-hacks/blob/master/dns-debugging/dumper.yaml

Assumption is that network-edge will clarify what to run to collect, and where it needs to run.

TRT will figure out how to get it running there during CI jobs, and how to collect the artifacts afterwards.