Unable to comission or deploy on two clusters that have been working for 2 years

So I have two MAAS stacks, each with 50 to 80 nodes. One stack is about a year older than the other, so one is running on ubuntu 18.04, and the other on 16.04.

Recently the deployments started failing on both stacks, hanging at “random: nonblocking pool is initiated”. both commissioning and deployments hang in the same spot, and it appears to be in the first boot from the ephemeral image so it blocks both commissioning and deployments.

I applied system updates, just in case my automatically updated images had drifted out of compliance with the MAAS API. The last system update for the MAAS systems was on September 10. I’m currently running MAAS version: 2.3.5 (6511-gf466fdb-0ubuntu1) and MAAS version: 2.4.2 (7034-g2f5deb8b8-0ubuntu1).

Any ideas how I can resolve this?

I found two cabling errors, creating an invalid network configuration. I think someone was reconfiguring a machine, and moved an uplink cable in error.