Confirming for anyone else searching. I got the above Glotrends card along with this m.2 adapter working with the DEG1:Yes I got it working and it is passthrough to a windows vm for gaming and ollama vm. I am assuming you have powered off your server first as it is not hot pluggable? What cards are you using?
I am using these:
GLOTRENDS PA09-HS M.2 NVMe to PCIe 4.0 X4 Adapter with M.2 Heatsink for 2280/2260/2242/2230 M.2 NVMe SSD
View attachment 40325ali
@anewsomeI was able to fit these 5mm heatsinks with thermal pad on my MS01. Gotta say, I think 5mm is the absolute tallest I could have fit in there. It's literally touching the fan bracket.
View attachment 37612
I bought the same exact heatsink based on his post and it fits (barely). Had no issues closing the case.@anewsome
I am slightly confused by this picture, especially because I also saw your rack photos with the 5x MS-01.
Could you still close the case with the heatsink in place, or did you remove them again?
There was an issue with early MS-01 units. I bought 2 units a few month after product launch. One was a barebone unit and the other came with RAM and SSD already installed. I repasted the CPU, put 96GB Crucial RAM and Samsung 990 Pro SSDs in the barebone unit, and it has beed working fine ever since. In the 2nd unit I just swapped the OEM RAM with 96GB Crucial, installed Proxmox, tested a few VMs, and left it running. When I came back the next day MS-01 was powered off. I pushed the power button and the blue LED flashlight came on, but MS-01 wouldn't POST no matter what. Swapping back to OEM RAM didn't help either, so I contacted Minisforum. They said that DDR training can take a long time, and I should let MS-01 boot for 2 hoursI have two MS-01s with I9-13900H CPUs and 96GB Crucial RAM. I had one of them drop out of the vCenter cluster and checking it found it was powered off. I powered it on and it wouldn't boot. I pulled the power cord and tried again without luck. Checking Hardware Information in AMT shows the two DIMMs but Event Log shows "No system memory is physically installed in the system."
Has this happened to anyone else before and if so what was the cause? I can't see both memory modules dying at the same time. I'm going to tear into it tomorrow and run more tests but wanted to check here as well.
why this is so cheap https://www.amazon.de/-/en/KALEA-IN...00-16e&qid=1737817535&sprefix=,aps,553&sr=8-1Looks like that HBA card consumes around 25W (or more) which is quite high. I guess you can't tell yet without the RAM but it would be interesting to hear what kind of temps you're getting with it. I'd be a bit concerned tbh. For reference the newer 9305-16e would be around 13W and the latest 9500-16e about 9W. I've ordered the Lenovo variant of the latter, 440-16e, which I was able to find for a somewhat digestible price. Won't arrive for a few weeks though, and apparently my MS-01 is delayed too.
The CMOS battery on one of mine failed, it's a royal pain to replace due to the location of the 2 pin connector.The same happened to me and i had to do an RMA.
Later i read somewhere, perhaps in forums, that possibly the cmos battery gets depleted and it is a case of buying one and replace it.
I think they are pretty cheap and its worth a try rather then waiting weeks for a replacement
Sorry if this was posted or not before but I find it hard to search for infos across this single ms-01 thread. I was thinking of removing the insulation around the heatsink of the cpu. I've already bought PTM7950 but wanted you advice before doing so with the insulationOk, guys I took one for the team lol. Don't even try to pull the adhesive cover off the cpu block it will tear mine did.
The paste is dried and cheap, it just flaked right off.
View attachment 35042
View attachment 35043
After I cleaned it good with alcohol:
View attachment 35046View attachment 35047
Liquid metal applied (Be careful you want a thin layer due to the pcb being so small and it took awhile of massaging to get the liquid metal to stick):
View attachment 35050View attachment 35051
Here is where the back screws are so you don't have to peal the adhesive cover:
View attachment 35052View attachment 35053
Idle temps I am seeing a 3C drop and I put the server under a load before and hit 65C and after with same test I hit 45C.
So end result 20C improvement.
Even if you don't do liquid metal I recommend new thermal paste.
Which CPU do you have? Bios shows it's running 4800 or whatnot, even though I set the max speed upward. Latest bios here. Maybe I am missing something? Here is my TrueNAS Scale output from using sudo dmidecode -t memoryI can confirm that Kingston Fury 64GB 5600 MHz works on MS-01
However, keep in mind it runs at 5200 MHz
I tried to install Google Coral Dual TPU in Wi-Fi slot (E Key), doesn't seem to be detected.
The B+M NVMe slot was detected however.
Also, running both SFP+ as bond works, in case someone else hasn't tried it.
```
Ethernet Channel Bonding Driver: v6.1.0-17-amd64
Bonding Mode: IEEE 802.3ad Dynamic link aggregation
Transmit Hash Policy: layer2 (0)
MII Status: up
MII Polling Interval (ms): 100
Up Delay (ms): 400
Down Delay (ms): 200
Peer Notification Delay (ms): 0
802.3ad info
LACP active: on
LACP rate: slow
Min links: 0
Aggregator selection policy (ad_select): stable
Slave Interface: enp2s0f1
MII Status: up
Speed: 10000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr:<redacted>
Slave queue ID: 0
Aggregator ID: 1
Actor Churn State: none
Partner Churn State: none
Actor Churned Count: 0
Partner Churned Count: 0
Slave Interface: enp2s0f0
MII Status: up
Speed: 10000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: <redacted>
Slave queue ID: 0
Aggregator ID: 1
Actor Churn State: none
Partner Churn State: none
Actor Churned Count: 0
Partner Churned Count: 0
```