A little update... I spent the better part of a work day today trying to figure out why the Chenbro refuses to recognize an Intel Optane 905P NVME/U.2 drive, and unfortunately have still not figured it out. Pretty frustrating honestly. It's quite possible my cabling from the nvme/u.2 backplane to the motherboard is incorrect, but I've tried what I believe to be every option available. Unfortunately, the user manual provided only covers cable connections for the -U08 (8 NVME U.2 slots) version and not the much more commonly available -U04 (4 NVME U.2 slots) version (the one OP and I both have). I was able to get the Optane drive to work perfectly, and with full perf, using a simple pcie to mini-sas card and mini-sas to U.2 cable with external power. The onboard mini-sas connectors designated for NVME/U.2 however, simply do not seem to want to work, regardless of ports used or bios settings changes made. Very odd, but I feel like if this was actually documented for this model (U04) this probably wouldn't be such an issue. I'm basically just stabbing in the dark currently. haha.
On top of that perplexing problem, I also discovered today that there are about 50 unknown devices in device manager (Windows server 2019), obviously due to lacking drivers. Now here's the kicker... ZERO drivers are available on the Chenbro site for this server model. Yup, ZERO. It's a shame, this is a really nice platform and enormous bang for the buck if not for these basic documentation and driver availability issues.... at least so far that I've seen. Makes me sad, and will be a big pain, but I'm starting to inch towards needing to RMA these three servers. Ouch.
EDIT: I contacted Chenbro's US support and got a pretty speedy response back from a very helpful person named Kevin. He forwarded along a driver pack that fixed all the issues I had above. Now I'm running into a bit of a roadblock trying to reduce the RPM thresholds on the four 80mm case fans via IPMI. Due to the way the thresholds are setup in the Bios/BMC, the fans run at a constant 8400 RPM. I issued the usual commands to edit them, but it appears they have them locked down. I sent a message to Kevin asking if he had any suggestions. The fans could easily run at half this speed, possibly a third, and still provide an abundance of cooling for my particular configuration, except perhaps in the heaviest loads (even then, probably still fine though). I'm using two 125w CPUs, and the system is spec'd for up to two 165w CPUs, so I imagine they went with the better safe than sorry approach rather than providing for a more gradual stepping up of the fans.
On Monday or Tuesday I'll begin testing the SFP and QSFP cards.