Topton Jasper Lake Quad i225V Mini PC Report

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

thimplicity

Member
Jan 11, 2022
61
9
8
Hi everyone,
Very nice thread. A pity that it developed after I ordered my N5105 box in May from AliExpress. I ordered it barebones and added 8GB RAM (Crucial CT8G4SFRA32A) and an NVMe (SK hynix Gold P31 500GB PCIe). I ran Memtest for 16h with no problems and then went ahead and installed proxmox and pfSense on it. It worked will, but after 7-10 I had the first random freezes and crashes.

I am troubleshooting for about a month now. I reinstalled pfSense with the recommended settings (I did some things wrong initially there), which helped a little. Then I thought it might be the temperatures. To I installed a heatsink on the NVMe and added a fan right on top of the unit. Not so much fanless anymore.

The CPU information looks OK to me, but I am no expert:

Code:
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   39 bits physical, 48 bits virtual
CPU(s):                          4
On-line CPU(s) list:             0-3
Thread(s) per core:              1
Core(s) per socket:              4
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       GenuineIntel
CPU family:                      6
Model:                           156
Model name:                      Intel(R) Celeron(R) N5105 @ 2.00GHz
Stepping:                        0
CPU MHz:                         2000.000
CPU max MHz:                     2900.0000
CPU min MHz:                     800.0000
BogoMIPS:                        3993.60
Virtualization:                  VT-x
L1d cache:                       128 KiB
L1i cache:                       128 KiB
L2 cache:                        1.5 MiB
L3 cache:                        4 MiB
NUMA node0 CPU(s):               0-3

The temperatures in proxmox look good as well:

Code:
coretemp-isa-0000
Adapter: ISA adapter
Package id 0:  +39.0°C  (high = +105.0°C, crit = +105.0°C)
Core 0:        +35.0°C  (high = +105.0°C, crit = +105.0°C)
Core 1:        +35.0°C  (high = +105.0°C, crit = +105.0°C)
Core 2:        +35.0°C  (high = +105.0°C, crit = +105.0°C)
Core 3:        +35.0°C  (high = +105.0°C, crit = +105.0°C)

acpitz-acpi-0
Adapter: ACPI interface
temp1:        +27.8°C  (crit = +119.0°C)

nvme-pci-0100
Adapter: PCI adapter
Composite:    +49.9°C  (low  =  -0.1°C, high = +82.8°C)
                       (crit = +83.8°C)
Sensor 1:     +42.9°C  (low  = -273.1°C, high = +65261.8°C)
Sensor 2:     +42.9°C  (low  = -273.1°C, high = +65261.8°C)
I know that I might have adopted to early and I am in discussions with AliExpress, because Topton started ghosting me (just like @dazagrt and @ReturnedSword) after some initial messages when they learned that I know my way around IT and hardware. I filed a claim with AliExpress now, but of course I am outside the return/refund window.

The weird thing is that proxmox itself runs very well, but pfSense randomly freezes/reboots.
It would be great to get some more ideas on how to get this thing stabilized.

Thanks in advance for some tips
 

burtal

Active Member
Jul 7, 2022
111
63
28
Yes, i understand, but try and test it to see if it happens again...this way you can isolate the issue.
Pfsense issue or proxmox issue...or both...or maybe a missconfig....troubleshooting would be easier with less variables in the enviroment
 
  • Like
Reactions: abq

thimplicity

Member
Jan 11, 2022
61
9
8
Yes, i understand, but try and test it to see if it happens again...this way you can isolate the issue.
Pfsense issue or proxmox issue...or both...or maybe a missconfig....troubleshooting would be easier with less variables in the enviroment
good point! I am using it as the productive router, so will need to plan a “family maintenance window”. What’s the best way to go about it?

- backup config
- Install pfsense baremetal
- Restore config

Should it be that easy?
 

gregg098

Member
Jun 7, 2019
38
25
8
Hi everyone,
Very nice thread. A pity that it developed after I ordered my N5105 box in May from AliExpress. I ordered it barebones and added 8GB RAM (Crucial CT8G4SFRA32A) and an NVMe (SK hynix Gold P31 500GB PCIe). I ran Memtest for 16h with no problems and then went ahead and installed proxmox and pfSense on it. It worked will, but after 7-10 I had the first random freezes and crashes.

I am troubleshooting for about a month now. I reinstalled pfSense with the recommended settings (I did some things wrong initially there), which helped a little. Then I thought it might be the temperatures. To I installed a heatsink on the NVMe and added a fan right on top of the unit. Not so much fanless anymore.

The CPU information looks OK to me, but I am no expert:

Code:
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   39 bits physical, 48 bits virtual
CPU(s):                          4
On-line CPU(s) list:             0-3
Thread(s) per core:              1
Core(s) per socket:              4
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       GenuineIntel
CPU family:                      6
Model:                           156
Model name:                      Intel(R) Celeron(R) N5105 @ 2.00GHz
Stepping:                        0
CPU MHz:                         2000.000
CPU max MHz:                     2900.0000
CPU min MHz:                     800.0000
BogoMIPS:                        3993.60
Virtualization:                  VT-x
L1d cache:                       128 KiB
L1i cache:                       128 KiB
L2 cache:                        1.5 MiB
L3 cache:                        4 MiB
NUMA node0 CPU(s):               0-3

The temperatures in proxmox look good as well:

Code:
coretemp-isa-0000
Adapter: ISA adapter
Package id 0:  +39.0°C  (high = +105.0°C, crit = +105.0°C)
Core 0:        +35.0°C  (high = +105.0°C, crit = +105.0°C)
Core 1:        +35.0°C  (high = +105.0°C, crit = +105.0°C)
Core 2:        +35.0°C  (high = +105.0°C, crit = +105.0°C)
Core 3:        +35.0°C  (high = +105.0°C, crit = +105.0°C)

acpitz-acpi-0
Adapter: ACPI interface
temp1:        +27.8°C  (crit = +119.0°C)

nvme-pci-0100
Adapter: PCI adapter
Composite:    +49.9°C  (low  =  -0.1°C, high = +82.8°C)
                       (crit = +83.8°C)
Sensor 1:     +42.9°C  (low  = -273.1°C, high = +65261.8°C)
Sensor 2:     +42.9°C  (low  = -273.1°C, high = +65261.8°C)
I know that I might have adopted to early and I am in discussions with AliExpress, because Topton started ghosting me (just like @dazagrt and @ReturnedSword) after some initial messages when they learned that I know my way around IT and hardware. I filed a claim with AliExpress now, but of course I am outside the return/refund window.

The weird thing is that proxmox itself runs very well, but pfSense randomly freezes/reboots.
It would be great to get some more ideas on how to get this thing stabilized.

Thanks in advance for some tips
Im pretty much in the same boat lately. I get a random crash that has no useful logs in Proxmox or OPNsense. Internet goes down until I log in and restart the VM. Might try standalone myself at some point if it keeps up. Also need to try passthrough NICs. Speed is great with Proxmox bridges though.
 

thimplicity

Member
Jan 11, 2022
61
9
8
Im pretty much in the same boat lately. I get a random crash that has no useful logs in Proxmox or OPNsense. Internet goes down until I log in and restart the VM. Might try standalone myself at some point if it keeps up. Also need to try passthrough NICs. Speed is great with Proxmox bridges though.
how often are your crashes?

I passed the NICs through before. No difference for me. Bare metal seems to be the next step and then bridge the time for a stable revision from AliExpress or Amazon
 

gregg098

Member
Jun 7, 2019
38
25
8
how often are your crashes?

I passed the NICs through before. No difference for me. Bare metal seems to be the next step and then bridge the time for a stable revision from AliExpress or Amazon
They're random. Was happening every night for a bit. Then I reinstalled OPNsense from scratch and tweaked a few things. Went a few days, then happened again. Tweaked a few more things, few more days, then happened yesterday morning.
 

gregg098

Member
Jun 7, 2019
38
25
8
Maybe someone here knows what this crash might be. N6005 version. OPNsense on Proxmox. No useful logs from OPNsense. VM shows a yellow paused symbol and says "internet-error" after this happens.

Jul 30 09:21:49 pxfirewall QEMU[1491]: KVM internal error. Suberror: 3
Jul 30 09:21:49 pxfirewall QEMU[1491]: extra data[0]: 0x0000000080000b0e
Jul 30 09:21:49 pxfirewall QEMU[1491]: extra data[1]: 0x0000000000000031
Jul 30 09:21:49 pxfirewall QEMU[1491]: extra data[2]: 0x0000000000000083
Jul 30 09:21:49 pxfirewall QEMU[1491]: extra data[3]: 0x0000000800a37fe0
Jul 30 09:21:49 pxfirewall QEMU[1491]: extra data[4]: 0x0000000000000001
Jul 30 09:21:49 pxfirewall QEMU[1491]: RAX=0000000800a37780 RBX=fffffe000cadc090 RCX=00000000c0000101 RDX=00000000ffffffff
Jul 30 09:21:49 pxfirewall QEMU[1491]: RSI=0000000000000000 RDI=fffffe000cadc090 RBP=fffffe000cadc080 RSP=fffffe000cadbfb0
Jul 30 09:21:49 pxfirewall QEMU[1491]: R8 =0000000000001001 R9 =00007fffffffea70 R10=0000000000000000 R11=00008700abfd19c8
Jul 30 09:21:49 pxfirewall QEMU[1491]: R12=0000000800a90240 R13=0000000800a1f000 R14=0000000000000013 R15=0000000800a37780
Jul 30 09:21:49 pxfirewall QEMU[1491]: RIP=ffffffff811251c1 RFL=00010082 [--S----] CPL=0 II=0 A20=1 SMM=0 HLT=0
Jul 30 09:21:49 pxfirewall QEMU[1491]: ES =003b 0000000000000000 ffffffff 00c0f300 DPL=3 DS [-WA]
Jul 30 09:21:49 pxfirewall QEMU[1491]: CS =0020 0000000000000000 ffffffff 00a09b00 DPL=0 CS64 [-RA]
Jul 30 09:21:49 pxfirewall QEMU[1491]: SS =0000 0000000000000000 ffffffff 00c00000
Jul 30 09:21:49 pxfirewall QEMU[1491]: DS =003b 0000000000000000 ffffffff 00c0f300 DPL=3 DS [-WA]
Jul 30 09:21:49 pxfirewall QEMU[1491]: FS =0013 0000000800287120 ffffffff 00c0f300 DPL=3 DS [-WA]
Jul 30 09:21:49 pxfirewall QEMU[1491]: GS =001b ffffffff82611000 ffffffff 00c0f300 DPL=3 DS [-WA]
Jul 30 09:21:49 pxfirewall QEMU[1491]: LDT=0000 0000000000000000 ffffffff 00c00000
Jul 30 09:21:49 pxfirewall QEMU[1491]: TR =0048 ffffffff82611384 00002068 00008b00 DPL=0 TSS64-busy
Jul 30 09:21:49 pxfirewall QEMU[1491]: GDT= ffffffff826113ec 00000067
Jul 30 09:21:49 pxfirewall QEMU[1491]: IDT= ffffffff81f5ba10 00000fff
Jul 30 09:21:49 pxfirewall QEMU[1491]: CR0=80050033 CR2=ffffffff811251c1 CR3=0000000800a37780 CR4=003506e0
Jul 30 09:21:49 pxfirewall QEMU[1491]: DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000
Jul 30 09:21:49 pxfirewall QEMU[1491]: DR6=00000000ffff0ff0 DR7=0000000000000400
Jul 30 09:21:49 pxfirewall QEMU[1491]: EFER=0000000000000d01
Jul 30 09:21:49 pxfirewall QEMU[1491]: Code=?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? <??> ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ??
 

dtw

New Member
Jul 11, 2022
14
6
3
Are this issues present on other OS ? like windows or a linux distro with GUI?
Debian 11 works reliably, Windows also seemed to work, but I didn't use it for a long time.

Also since you are in testing phase .... you can try the latest kernel .... maybe it solves something

Thank you for the pointer! I did install it but sadly only saw a small change, which is that ACPI error messages I saw on boot have disappeared (see screenshots); the machine does still hang promptly if booted without `nomodeset`, and within minutes (which is longer than it took before) when booted headless with `nomodeset`.

As for the BIOS update, I did start off by installing the v2 BIOS I found on the Changwang website before doing anything else.... I'm now thinking I should have backed up my original BIOS first so I could go back. @yxman would you be willing to dump the bios from your MW-JSL2.5G-4L? You can do this e.g. by running AfuEFIx64 from an EFI Shell. I've crated this efiboot image to make that as easy as possible. You can flash it onto a USB drive and boot from it to get an EFI shell. On the EFI Shell enter:
Code:
fs1:
AfuEFIx64 backup.bin /o
You can then reboot with "reset".

I used the same mechanism to flash Changwang's BIOS like so: "AfuEFIx64 MJS4LT11-HDMI-2.bin /P /B /N /X" as recommended by Changwang, which in retrospect I should not have done without creating a backup first...

Maybe your memory is unstable/faulty... check with passmark memtest

Thank you for the suggestion; I ran it with all tests for 48 hours and while the little box got hot, it found no errors.

not yet, as I would like to use the box for more than just pfSense. So I tried to stick with proxmox until now.
If you can at all use a different router for now I'd suggest installing Debian 11 and seeing how it does. It works like a charm for me and would help narrow down the issue. I'm highly suspicious it's the i915 graphics which seem to be causing issues. xcp-ng also couldn't get a console running on dom0 without the i915 driver, which seems unusual, though I'm not familiar at all with Xen (Unable to install XCP-ng on some hardware with i915 compatible graphics card · Issue #436 · xcp-ng/xcp).

In the case of xcp-ng there's no crash though, and with the i915 and i225 drivers from the iso linked from the above github issue it works fine but with roughly double the power consumption of Debian 11 headless, which I'm not enjoying.

I'll try some more to narrow down what is actually going wrong with the proxmox kernel and might try and file a report at Bugzilla Main Page if I can get any insights. So far it seems to lock up without any kernel messages but I'll poke around some more.
 

Attachments

RonG

New Member
Jul 31, 2022
10
2
3
I have the x4c topton w n5105
Can this handle 64gb ram? I see the cpu claims 16gb max from Intel but users are putting in 32gb(2x16gb) so-dimms.
What are the requirements for 32gb or 64gb? R1x8 single ranked sodimms or R2x8 or even R1x16? Most posts state a brand but not a specific model of 16 or 32gb sodimms to use. Just curious if I can get 64gb on this puppy?
 

RonG

New Member
Jul 31, 2022
10
2
3
They're random. Was happening every night for a bit. Then I reinstalled OPNsense from scratch and tweaked a few things. Went a few days, then happened again. Tweaked a few more things, few more days, then happened yesterday morning.
Watchdog is disabled in bios yes?
 

dtw

New Member
Jul 11, 2022
14
6
3
I have the x4c topton w n5105
Can this handle 64gb ram? I see the cpu claims 16gb max from Intel but users are putting in 32gb(2x16gb) so-dimms.
What are the requirements for 32gb or 64gb? R1x8 single ranked sodimms or R2x8 or even R1x16? Most posts state a brand but not a specific model of 16 or 32gb sodimms to use. Just curious if I can get 64gb on this puppy?
I'm not sure if anyone has reported on 64GB, but for 32GB we know 2Rx8 and 1Rx8 both work:

I forgot, we already know that 2x 16GB dual-ranked (2Rx8) memory kits work on these Jasper Lake devices, but I wanted to confirm that 2x 16GB single-rank (1Rx8) version of Crucial's CT16G4SFRA32A also work. This module can come as 1Rx16, 1Rx8, or 2Rx8 and has the same product number. I purchased the 32GB kit (CT2K8G4SFRA32A) and tested it for 24h in memtest86 with no errors on my N5105 running at 2933MT. I haven't seen anyone confirm that 16GB 1Rx16 modules (= with 4 chips only) work, but both 2Rx8 as 1Rx8 work fine based on my tests.
 

RonG

New Member
Jul 31, 2022
10
2
3
I need to check. Can you explain what this does? And it should be disabled or enabled?

Asking because THIS thread makes the statement

Thanks.
Not an expert but it's a hardware watchdog that talks to the OS and if not response after (seconds, minutes, can be set), it triggers a reboot of the hardware. My board had it on by default but proxmox 7.2.7 wasn't setup for a watchdog and the system kept rebooting hourly. Watchdog is good to have on so it can reboot a locked machine, but the OS needs to support it. Linux does but may need to be configured to work with bios settings of watchdog.
 

gregg098

Member
Jun 7, 2019
38
25
8
Not an expert but it's a hardware watchdog that talks to the OS and if not response after (seconds, minutes, can be set), it triggers a reboot of the hardware. My board had it on by default but proxmox 7.2.7 wasn't setup for a watchdog and the system kept rebooting hourly. Watchdog is good to have on so it can reboot a locked machine, but the OS needs to support it. Linux does but may need to be configured to work with bios settings of watchdog.
Thanks. I will have to check the setting, but Proxmox is running rock solid. Just OPNsense dies. Same VM that ran great in another machine that I tested with before I got the Topton box. Even rebuilt the VM from scratch with slightly different settings and same crash.

I just created, and am still testing, a script that checks the VM status with pvesh every 30 seconds and restarts it if its stopped. Running it as a service so I can disable it for updates and maintenance. Works so far on a dummy vm, but I'm just stopping it with
Code:
qm stop VMID
and not actually seeing the vm in a crash state, so I'm not sure how itll handle it. If it works, Ill post the code for everyone.
 

thimplicity

Member
Jan 11, 2022
61
9
8
Thanks. I will have to check the setting, but Proxmox is running rock solid. Just OPNsense dies. Same VM that ran great in another machine that I tested with before I got the Topton box. Even rebuilt the VM from scratch with slightly different settings and same crash.

I just created, and am still testing, a script that checks the VM status with pvesh every 30 seconds and restarts it if its stopped. Running it as a service so I can disable it for updates and maintenance. Works so far on a dummy vm, but I'm just stopping it with
Code:
qm stop VMID
and not actually seeing the vm in a crash state, so I'm not sure how itll handle it. If it works, Ill post the code for everyone.
We really seem to be in the same boat here, just opensense vs pfsense. I will need to check the watchdog setting as well.

Did you update the bios already? I build computers since many years, but I never updated the BIOS in my life. I am especially cautious because of this Chinese box.

I would be interested in the script as well, but for me it is more reboots than freezes at the moment. Although I am coming up on 2 days without a problem. Small victories! It is amazing how important a stable router is, but you will only appreciate it when it is not stable.
 

gregg098

Member
Jun 7, 2019
38
25
8
We really seem to be in the same boat here, just opensense vs pfsense. I will need to check the watchdog setting as well.

Did you update the bios already? I build computers since many years, but I never updated the BIOS in my life. I am especially cautious because of this Chinese box.

I would be interested in the script as well, but for me it is more reboots than freezes at the moment. Although I am coming up on 2 days without a problem. Small victories! It is amazing how important a stable router is, but you will only appreciate it when it is not stable.
I did try some of the tweaks found in this thread, but I ended up resetting to default after some of the problems I had. I dont think either made a difference though. I did not update the BIOS.