You have a CAS (stupid; don’t do it unless you have 100K site or more!)
You have a CM07 hierarchy and a CM12 hierarchy both running at the same time managing clients. Move all subnets out of CM07 and into CM12 marked for boundary group client assignment so that all clients set to AUTO will fail in CM07 (so you hard code client assignments staying on CM07) use AUTO in CM12 (so clients will assign automatically to the correct site).
Some CM07 clients show 2 MPs in their locationservices.log and show CM12 DPs listed in the locationservices.log and the CAS.log. Clients are stuck in waiting for content status as they try to reach CM12 DPs.
Make sure that you didn’t accidentally set the DPs for assignment as well under the references tab of your primary site assignment properties for your boundary groups. If they show for content location, remove them.
With no content set, clients also need to fallback to any DP. This works great for those of us using 1E Nomad, but other folks could experience cross WAN traffic they don’t want. We saw only 2-3% of clients find a CM12 MP like this. Most were fine and still worked fine. So you might want to just ride it out; fail forward by installing the CM12 client on machines showing the issue. We chose to rip off content location on the boundary group because 1E saves us WAN worries.
For folks on Nomad with a very flat hierarchy, consider leaving things setup without boundary groups set for content. You’ll have more DPs at your disposal for a greater chance of uptime and resiliency.