Minisforum MS-01 PCIe Card and RAM Compatibility Thread

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Stealthy

New Member
Sep 27, 2024
8
6
3
Do we have to wait for a BIOS update to apply the 0x12B Microcode that fixes the 13th gen Intel CPU problems? Or is that something we can apply outside of the BIOS?
The 13th Gen low power mobile CPUs (which are in the MS-01 computers) are not affected like the normal 13th gen intel CPU's.

So the Microcode does nothing for the MS-01.

"Based on our in-depth analysis of the reported Intel Core 13/14 Gen desktop processor instability issues, Intel has determined that mobile products are not exposed to the same issue. The symptoms being reported on 13/14 Gen mobile systems – including system hangs and crashes – are common symptoms stemming from a broad range of potential software and hardware issues.
 
  • Like
Reactions: Avamander and P00r

ViciousXUSMC

Active Member
Nov 27, 2016
277
147
43
41
Why your Cpu MhZ are so low? you have the defaults in the bios as matter CPU voltages right?
Anyway, i had delete the voltage numbers in the bios as a person suggested in this thread in the early days and have no problem reaching average 3642mhz from all cores of my 12900h with 96gb, Bios 1.17 and most cores under 80c. The above with PTM7950 that i went to it lately. Before i had Thermal Grizzly Conductonaut Extreme were after 2-3 months the temperatures raised at 90c+.
Ambient is at 28c atm and s-tui is running more than 10mins.
View attachment 38867
Cryo is a great compound for performance but it has a very short life span, I learned this the hard way.
Last PC build I did I went back to tried and true stuff.

Something like MX-4, NT-H2 etc.

I dont know if I have participated much in this thread.
But I had an issue where my current MS-01 didnt want to power on after power loss a couple weeks ago (hurricane) but I got it back to life by taking the cover off for some reason and hitting the power button directly (that cant have been the fix, so it must have been something with just waiting long enough for it to turn on again)

This leaves me with less confidence about this system but I cant help feel its perfect for my needs so I just ordered another one.

The first one was the 12900H with the 1TB 32GB combo and I added a 10gb RJ45 interface so I can use 10gb WAN service from my ISP and its running PFSense.

The one I just ordered I got barebone and the 96GB RAM kit on the first post as its on sale right now, and I am going to install either 2x 1TB or 2x 2TB Samsung 990 Pro and run them in Raid 1 and install Proxmox.

I plan on finally turning off my old Dell R710 and retire it as my old NAS and Hypervisor.
Electricity is going up and up and up and I want to see if I can start minimizing my rack that is always on 24/7 to save some power.

Just you know purchases like this for $800 it likely would take a decade to save that much on power, its hard to justify on savings alone.
 
  • Like
Reactions: SteffenAL

koubi

New Member
Sep 10, 2024
6
3
3
Stability test on going :

* 1 frigate instance (debian LXC + docker) with coral TPU /dev passthrough and 1 camera stream (dummy, just to get some activity)
* 1 rocky linux VM that generates 10GB of random data, and uploads it to another server, then downloads it back; every hour
* 1 windows 11 VM, idling on the Edge default new tab page (you know, with all the ads and news that update from time to time).

I put system on UPS to get some resilience against power grid instability.

So far, no issue for 3 days, system idling at 24W, CPU right below 40°C, silent.

1728631383539.png

1728631395461.png

1728631435357.png

Recalling config : barebone i5-12600H, 2*16GB Crucial (CT2K16G56C46S5), 256GB M.2 Patriot SSD (yep, still no 4TBs SSDs), system is connected via 1 of the SFP ports to my LAN with 1 passive twinax from 10gtek (CAB-10GSFP-P2M). BIOS is 1.22
 

koubi

New Member
Sep 10, 2024
6
3
3
  • Like
Reactions: webmaster

webmaster

New Member
Jan 26, 2024
27
3
3
The iGPU should be enough ? or do you need 1 discret GPU as well ?
Thanks!

It is running in a VM, and my understanding is that it is not possible to get video out with the iGPU (Iris XE)?

Therefore the idea to get a discrete GPU that can be passed through.
 

Jarery

New Member
Mar 7, 2024
5
6
3
No reboots so far. 7 days uptime running proxmox with P+E cores and HT enabled, that’s a first for me.
BIOS 1.22
Microcode 00004122
Kernel 6.8.12-2-pve
is there anything you updated or changed that you attribute to the stability?
is the microcode what came with the system along with bios 1.22 ?
 

BlueChris

Active Member
Jul 18, 2021
154
54
28
53
Athens-Greece
Guys, 3 days ago i got a sudden crash after a huge time with super stability.
In the crash log i saw this type of entries

Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: I/O tag 578 (1242) opcode 0x2 (I/O Cmd) QID 4 timeout, aborting req_op:READ(0) size:131072
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: I/O tag 705 (72c1) opcode 0x2 (I/O Cmd) QID 5 timeout, aborting req_op:READ(0) size:131072
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: I/O tag 706 (12c2) opcode 0x2 (I/O Cmd) QID 5 timeout, aborting req_op:READ(0) size:131072
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: I/O tag 707 (12c3) opcode 0x2 (I/O Cmd) QID 5 timeout, aborting req_op:READ(0) size:131072
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: I/O tag 708 (32c4) opcode 0x2 (I/O Cmd) QID 5 timeout, aborting req_op:READ(0) size:40960
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: I/O tag 695 (42b7) opcode 0x2 (I/O Cmd) QID 9 timeout, aborting req_op:READ(0) size:131072
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: I/O tag 313 (b139) opcode 0x2 (I/O Cmd) QID 11 timeout, aborting req_op:READ(0) size:126976
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: I/O tag 315 (113b) opcode 0x2 (I/O Cmd) QID 11 timeout, aborting req_op:READ(0) size:131072
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:13 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:16 BCProxmox kernel: nvme nvme0: I/O tag 317 (e13d) opcode 0x1 (I/O Cmd) QID 11 timeout, aborting req_op:WRITE(1) size:12288
Oct 11 04:12:16 BCProxmox kernel: nvme nvme0: I/O tag 912 (7390) opcode 0x1 (I/O Cmd) QID 13 timeout, aborting req_op:WRITE(1) size:40960
Oct 11 04:12:16 BCProxmox kernel: nvme nvme0: Abort status: 0x0
Oct 11 04:12:16 BCProxmox kernel: nvme nvme0: Abort status: 0x0
and after that, this errors appear

Oct 11 04:12:43 BCProxmox kernel: nvme nvme0: I/O tag 578 (1242) opcode 0x2 (I/O Cmd) QID 4 timeout, reset controller
Oct 11 04:12:44 BCProxmox pvestatd[1986]: VM 102 qmp command failed - VM 102 qmp command 'query-proxmox-support' failed - got timeout
Oct 11 04:12:44 BCProxmox pvestatd[1986]: status update time (8.267 seconds)
Oct 11 04:12:54 BCProxmox pvestatd[1986]: VM 102 qmp command failed - VM 102 qmp command 'query-proxmox-support' failed - unable to connect to VM 102 qmp socket - timeout after 51 retries

nvme0 is a Samsung 2TB 990 Pro that has all my VMs in proxmox with ZFS.
In my searching i spotted some people having problems with this in linux and Samsuns SSDs and the solution that they did was to put this extra line in grub to disable the Power Save features of the disk (APST)

nvme_core.default_ps_max_latency_us=0
Some info on APST
default_ps_max_latency_us = default power state max latency (microseconds)

Users can set ps_max_latency_us to zero to turn off APST

It tells the maximum allowed time for waking up from APST power management states is exactly 0 microseconds, which causes the APST feature to be disabled.

So when set to 0, the SSD won't enter power management states autonomously, which means it should remain operational and not enter any power-saving modes
Now again all seem fine and i only got +3 to +4 watts more power consumption from the machine but i am fine with that. (i got a 200gb Optane, the Samsung 990 Pro 2TB and a Samsung PM983a 8TB)
 
  • Like
Reactions: ViciousXUSMC

ViciousXUSMC

Active Member
Nov 27, 2016
277
147
43
41
Guys, 3 days ago i got a sudden crash after a huge time with super stability.
In the crash log i saw this type of entries



and after that, this errors appear




nvme0 is a Samsung 2TB 990 Pro that has all my VMs in proxmox with ZFS.
In my searching i spotted some people having problems with this in linux and Samsuns SSDs and the solution that they did was to put this extra line in grub to disable the Power Save features of the disk (APST)



Some info on APST


Now again all seem fine and i only got +3 to +4 watts more power consumption from the machine but i am fine with that. (i got a 200gb Optane, the Samsung 990 Pro 2TB and a Samsung PM983a 8TB)
This might be super important since I am building a Proxmox server with Samsung 2TB 990 Pro.
I would hate for my new "server" to be unstable when my R710 has been rock solid stable for half a lifetime.
 

Maunose

New Member
Feb 29, 2024
15
15
3
is there anything you updated or changed that you attribute to the stability?
is the microcode what came with the system along with bios 1.22 ?
I updated pretty much everything, including the MS-01 itself since I had to RMA my older unit. When the new unit arrived, it was already loaded with BIOS v1.22 but I updated Intel microcode using the files on github, and updated all proxmox packages using apt-get.
I have a strong feeling that this new unit may be a newer hardware revision that maybe fixed the instability issues, but I can't say that for certain as too much was update on the software side as well.
My machine has currently over 2 weeks of uptime and I'm even running a Win10 VM 24/7 for over a week (older machine would restart within minutes when running windows VMs).
 

enchanted_dragonfly

New Member
Oct 23, 2024
4
1
3
I've got an MS-01 running ProxMox and I have been following this thread since before I bought it. First post says that the NICGIGA AQC113 works. I got this one from Amazon.

I installed it, but can't quite get it working.

I see this from lspci -v:
Code:
01:00.0 Ethernet controller: Aquantia Corp. AQtion AQC113 NBase-T/IEEE 802.3an Ethernet Controller [Antigua 10G] (rev 03)
    Subsystem: Aquantia Corp. AQtion AQC113 NBase-T/IEEE 802.3an Ethernet Controller [Antigua 10G]
    Flags: fast devsel, IRQ 16, IOMMU group 15
    Memory at 6c800000 (64-bit, non-prefetchable) [size=512K]
    Memory at 6c8a0000 (64-bit, non-prefetchable) [size=4K]
    Memory at 6c400000 (64-bit, non-prefetchable) [size=4M]
    Expansion ROM at 6c880000 [disabled] [size=128K]
    Capabilities: [40] <chain broken>
    Kernel modules: atlantic
But it doesn't show up in ip a:
Code:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host noprefixroute
       valid_lft forever preferred_lft forever
2: enp88s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master vmbr0 state UP group default qlen 1000
    link/ether 58:47:ca:77:c6:5c brd ff:ff:ff:ff:ff:ff
3: enp91s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 58:47:ca:77:c6:5d brd ff:ff:ff:ff:ff:ff
4: enp3s0f0np0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 58:47:ca:77:c6:5a brd ff:ff:ff:ff:ff:ff
5: enp3s0f1np1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 58:47:ca:77:c6:5b brd ff:ff:ff:ff:ff:ff
6: vmbr0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 58:47:ca:77:c6:5c brd ff:ff:ff:ff:ff:ff
    inet 192.168.51.210/24 scope global vmbr0
       valid_lft forever preferred_lft forever
    inet6 fe80::5a47:caff:fe77:c65c/64 scope link
       valid_lft forever preferred_lft forever

And dmesg | grep atlantic shows me this:
Code:
[    2.022115] atlantic: FW Restart failed
[    2.022235] atlantic: probe of 0000:01:00.0 failed with error -5
I'm fairly new to ProxMox as the MS-01 is the first system I've installed it on.

Anyone know what I'm doing wrong?
 

iru

New Member
Oct 24, 2024
3
1
3
Hey yo I just pulled the trigger and ordered the 12600H version. Happy to join the conversation and hoping I won't be adding to the horror stories related to their support.

Got 96GB aftermarket RAM. Will be transferring my existing installation currently running on an HP mini 705 G4. It's running NixOS and nixos-containers (systemd-nspawn style containers) for some web applications, file server and other loads. No VMs for now but I might add something later.

I don't have high CPU needs but I want the IO and extension capabilities on that small packet so I can continue using my 10" rack. The rationale I told myself for this purchase is that this should last for at least 3 or 5 years.
 
  • Like
Reactions: Jaymac

Jaymac

New Member
Oct 23, 2024
5
1
3
Hey yo I just pulled the trigger and ordered the 12600H version. Happy to join the conversation and hoping I won't be adding to the horror stories related to their support.

Got 96GB aftermarket RAM. Will be transferring my existing installation currently running on an HP mini 705 G4. It's running NixOS and nixos-containers (systemd-nspawn style containers) for some web applications, file server and other loads. No VMs for now but I might add something later.

I don't have high CPU needs but I want the IO and extension capabilities on that small packet so I can continue using my 10" rack. The rationale I told myself for this purchase is that this should last for at least 3 or 5 years.
Hello Iru
I just ordered one myself yesterday as well. (UK)
I ordered the barebones 13900H, & also ordered 64gb crucial ram, 1 x 2gb wd 850x m.2 for the operating system etc, 2 x 4gb wd 850x m.2 for some remote fast storage, and a sfp+ cable to directly connect to my router with sfp+ port.

I have no clue what operating system I shall go with yet.

Jay
 

enchanted_dragonfly

New Member
Oct 23, 2024
4
1
3
I'm having trouble updating the BIOS.

I downloaded the ZIP, formatted a USB drive FAT32, unzipped the files to the USB.

On booting I hit F7 and pick UEFI OS, but then I get a blue window which is counting down. When it hits 0, it reboots normally. If I press a key to stop the countdown, then I get some options and if I say "Continue booting" it reboots normally. There is an option for "Reset" but I'm afraid to pick that.

How do you successfully get into UEFI shell?
 

mgolliday

New Member
Nov 14, 2022
3
0
1
I'm having trouble updating the BIOS.

I downloaded the ZIP, formatted a USB drive FAT32, unzipped the files to the USB.

On booting I hit F7 and pick UEFI OS, but then I get a blue window which is counting down. When it hits 0, it reboots normally. If I press a key to stop the countdown, then I get some options and if I say "Continue booting" it reboots normally. There is an option for "Reset" but I'm afraid to pick that.

How do you successfully get into UEFI shell?
 

mgolliday

New Member
Nov 14, 2022
3
0
1
You just need a USB that can actually boot into UEFI. I've never had to cold boot my MS-01 - installed both 1.25 and 1.26 this way