Minis forum MS-A2
- Thread starter freegate
- Start date
-
- Tags
- minisforum ms-01 ms-a2
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Can you please show a picture on how that fan was mounted? the system fan, I see that you are now longer using the support to hold it.With LM I idle at 38C and under load never go over 60C. I have opened the case up and additional time to apply more LM because the Copper is absorbing some of it.
View attachment 45204
View attachment 45205
using nvidia-smi --persistence-mode=1
View attachment 45206
using my 2 drive nvme mount https://forums.servethehome.com/index.php?threads/minis-forum-ms-a2.47735/post-474062 which just repurposes the stock nvme fan and heatsink.
Using this case mod as well Minisforum MS A2 / MS 01 case by Kevin68 | Download free STL model | Printables.com
View attachment 45208
View attachment 45209
View attachment 45210
Using this modded case was 1-2C hotter than running with no shell which is phenomenal.
Ambient is 70F
Using a custom 3d printed mount
The only thing that keeps multiple drives cool. the stock support 2 of the 3 drives gets way too hot.
The only thing that keeps multiple drives cool. the stock support 2 of the 3 drives gets way too hot.
so you just keep it standing up like that on the desk ?Using a custom 3d printed mount
The only thing that keeps multiple drives cool. the stock support 2 of the 3 drives gets way too hot.
Do you by any chance have rack-mountable version ?
I've had an ongoing issue with using LSI SAS3216/9300-series controllers in my MS-A2.
Hardware:
2 x MS-A2 on the latest BIOS
Custom 3d Printed case with a 120mm fan attached, which blows directly on the heatsink of the LSI controller.
LSI controllers: 9305-16e & 9300-8e
External ICY DOCK MB998IP-B 8-Bay enclosure
SM 220 power supply to power the enclosure
8 x Samsung 3.84TB SATA SSDs
OS:
The latest TrueNAS CE edition
ZFS RAID Z1 x 4 x 2 Wide
Issue:
The drive(s) in my external enclosure show as unavailable or needing replacement after a random period of time. I can remove the drive, replace it with another (I have several extra drives), and resliver, but that only temporarily gets the pool back into a good state until the next drive falls out of the array.
I have cycled through all of my drives at one point or another and have used these in other systems, so I know they're good. I have replaced the SFF-8644 cables, replaced the external enclosure with another, and swapped the 9305 controller for another 9300-8e, and the same issues persist.
To rule out my hardware, I've taken the HBA, cables, and the enclosure with the identical drives, stuck them in my spare SuperMicro server, and installed TrueNAS CE on it. It's run for weeks without any issues at all. I have two MS-A2s and was really hoping to use them for a new NAS setup, but I'm pulling my hair out trying to figure out what's going on. Nothing makes sense at this point other than it being the MS-A2s. My temperatures look good as well due to the 120 mm fan running.
Does anyone have any ideas or similar experience running LSI HBAs on these machines? Any input you can give me would be helpful at this point. I want to email support and see what they have to say.
Temperatures:
CPU - 41
ROC on LSI card: 40 (this is even under a full scrub of all drives)
Hardware:
2 x MS-A2 on the latest BIOS
Custom 3d Printed case with a 120mm fan attached, which blows directly on the heatsink of the LSI controller.
LSI controllers: 9305-16e & 9300-8e
External ICY DOCK MB998IP-B 8-Bay enclosure
SM 220 power supply to power the enclosure
8 x Samsung 3.84TB SATA SSDs
OS:
The latest TrueNAS CE edition
ZFS RAID Z1 x 4 x 2 Wide
Issue:
The drive(s) in my external enclosure show as unavailable or needing replacement after a random period of time. I can remove the drive, replace it with another (I have several extra drives), and resliver, but that only temporarily gets the pool back into a good state until the next drive falls out of the array.
I have cycled through all of my drives at one point or another and have used these in other systems, so I know they're good. I have replaced the SFF-8644 cables, replaced the external enclosure with another, and swapped the 9305 controller for another 9300-8e, and the same issues persist.
To rule out my hardware, I've taken the HBA, cables, and the enclosure with the identical drives, stuck them in my spare SuperMicro server, and installed TrueNAS CE on it. It's run for weeks without any issues at all. I have two MS-A2s and was really hoping to use them for a new NAS setup, but I'm pulling my hair out trying to figure out what's going on. Nothing makes sense at this point other than it being the MS-A2s. My temperatures look good as well due to the 120 mm fan running.
Does anyone have any ideas or similar experience running LSI HBAs on these machines? Any input you can give me would be helpful at this point. I want to email support and see what they have to say.
Temperatures:
CPU - 41
ROC on LSI card: 40 (this is even under a full scrub of all drives)
Last edited:
I have proxmox issues with kernel panic since last update on 6.17 kernel with MS-A2. Is anyone having troubles also?
Yes!!!!!I have proxmox issues with kernel panic since last update on 6.17 kernel with MS-A2. Is anyone having troubles also?
Was having the same type of issue, every ~12 hours or less, my MS-A2 would reboot. I think it was a kernel panic though the logging wasn't clear.
I went back to 6.14.8-2-pve and that seems to have fixed it; I believe a newer 6.14 kernel also had the same problem which makes me think whatever the problem was backported to later 6.14 kernels.
If you want to pursue this in the Proxmox forums, I would encourage you to. I just don't really have the time to experiment...
No Problems at all here with the 9955hx version. Had a 30 Days uptime before updating and rebooting.I have proxmox issues with kernel panic since last update on 6.17 kernel with MS-A2. Is anyone having troubles also?
I only changed these settings:
- Advanced->AMD CBS->SMU Common Options->TjMAX to 78
- PowerLimit Setting to Balance Mode
- Fan Mode to Balance
- all SSD Lane Speeds to Gen 4
- UMA Frame Buffer Size 1g
- Disabled Wifi and Audio
- Ac Loss Control to Always On
Everything else on Auto/Default.
- Advanced->AMD CBS->SMU Common Options->TjMAX to 78
- PowerLimit Setting to Balance Mode
- Fan Mode to Balance
- all SSD Lane Speeds to Gen 4
- UMA Frame Buffer Size 1g
- Disabled Wifi and Audio
- Ac Loss Control to Always On
Everything else on Auto/Default.
How much RAM do you have?No Problems at all here with the 9955hx version. Had a 30 Days uptime before updating and rebooting.
Using the Crucial 128GB CT2K64G56C46S5 KitHow much RAM do you have?
Okay, so... we have the same RAM... so that rules out my theory that perhaps the newer kernels were unhappy with 128GB...Using the Crucial 128GB CT2K64G56C46S5 Kit
How are your temps? Does memtest86 run without errors for 12h+?Okay, so... we have the same RAM... so that rules out my theory that perhaps the newer kernels were unhappy with 128GB...
I have not tried running memtest86 for 12+ hours I don't think, but I would note that mine is rock solid in 6.14.8-2 and every previous kernel... makes me think hardware issue is unlikely...How are your temps? Does memtest86 run without errors for 12h+?
Nothing that I can think of...Weird, any special configurations or pcie/other hardware passthrough? Any error logs?
I couldn't find much in the way of logs back when this was happening, and honestly, I just went back to the older kernel and left it at that. A system that spontaneously reboots with running VMs is worse than useless... and I was really hoping others would report the issue...
I'm using 64gb ram on my side, crucial. I changes bios params:
CPU Common Options :
- Core Performance Boost : Disabled # PBO off (stabilité, freq 2.5-4.3GHz)
- Global C-state Control : Disabled # No idle panics (vs Auto)
- SVM Mode : Enabled # Virt OK
- SVM Lock : Disabled # no lock
Since this change it seems to be working on the last kernel. I had the fans go on and off often before, now it's quiet. But with these settings I loose some power of the CPU.
CPU Common Options :
- Core Performance Boost : Disabled # PBO off (stabilité, freq 2.5-4.3GHz)
- Global C-state Control : Disabled # No idle panics (vs Auto)
- SVM Mode : Enabled # Virt OK
- SVM Lock : Disabled # no lock
Since this change it seems to be working on the last kernel. I had the fans go on and off often before, now it's quiet. But with these settings I loose some power of the CPU.