My refund was issued, so either the supplier has verified the RAM defective or refunded regardless. I've not seen any reports of issues with the Kingston Impact Fury Memory on this thread from what I recall. I personally would buy Kingston over Crucial if I buy another one of these units.
CWWK/Topton/... Nxxx quad NIC router
- Thread starter Becks0815
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Hi,
at this point I think my posts are just noise and do not provide much useful information; in any case :
after my previous post, I removed all covers from the device and only left RAM installed, attached a new PSU, strapped a 12cm fan on top pulling air, turned the device on its side and on the bottom side I placed a small desktop fan blowing air. Around 8 hours later, memtest86 finished succesfully (all 4 passes). A few hours later I restarted the test (didn't even touch the device) and it rebooted during the 4th pass.
Highest RAM temperature I've seen during this whole time was 51c/124f (temp stayed under 45c/113f most of the time). I don't think I can ask for lower temperatures during a stress test.
Then I decided to install one m2 and install windows 10 (since there is a plethora of stress/burn in tests for windows). I fully updated windows and run an old passmark burn in test utility - sometimes it finishes with no errors, other times it fails with errors and one time it rebooted during the test. Again, I barely touch/move the device, I just wait for a bit for it to cool off and restart the test. At no point does it feel uncomfortably hot and the new psu barely feels warm. note : during these tests I had a cable attached to ETH0 and a mouse along with the RAM and m2 (and in all cases/test I also have monitor and keyboard attached).
At this point I think I have to give up or buy another RAM stick. I've already contacted my supplier, will send them my current RAM stick for them to test but will probably take at least a week for them to get back to me. If they say the RAM is fine I'll probably buy a 16GB ram module that someone else suggested in this thread.
I've already ordered a NOCTUA NF-F12 INDUSTRIALPPC-2000 120MM PWM, should fit nicely on top of the device and I think I can screw it on (albeit scratching the case/heatsink) and an 8cm fan for the bottom (there seem to be screw holes for 8cm and 6cm fans). I think I can connect them directly to the motherboard but for now I'll just connect them to a separate small psu. I'm pretty sure at this point that 'fanless' is just a gimmick, I'm wondering how many people actually use these devices without a fan of some sort ...
Some notes of interest :
1. I have no motherboard drivers from CWWK, I had to download drivers for the i226 from intel and fully update windows hoping that windows update would download the appropriate drivers.
2. I cannot read CPU temperature in windows 10 (tried open hardware monitor, speccy etc). I can however read m2 temperatures and the highest it got was 62c/143f. I didn't spend much time trying to figure this out.
3. For a couple of days, every time I started/rebooted the device (either by removing power cable or ctrl-alt-del), there was a very long beep (I timed it at ~14sec, yes fourteen seconds). When the beep starts, monitor exits sleep mode but is blank, when the beep stops I see post screen and the device boots normally. What's up with that ? Can't find what the beep means in any bios error code list ... Now it does this occasionally but can't figure out why (it can happen after a failed test but can also happen after just powering on the device).
4. EDIT : forgot to mention that I tried to copy files to/from the device and only got around 550-600 mbps, it could be the drivers I found on intels site. ETH0 is connected to a 1gbps switch, every other device on that switch can do proper 1gbps apart from this one ... I didn't bother connecting it to a 2.5gbps switch yet.
As expected, no answer from CWWK. I don't know if they're a real company or just a couple of guys in a warehouse packaging/rebranding hardware they bought from somewhere else. I used the contact form provided on their webpage, is there other means of reaching them ? Perhaps a 'proper' BIOS with more options would help me sort this out.
I don't know what else to try. Anyways, I'll provide another update if I ever figure this out, thank you all for your time !
at this point I think my posts are just noise and do not provide much useful information; in any case :
after my previous post, I removed all covers from the device and only left RAM installed, attached a new PSU, strapped a 12cm fan on top pulling air, turned the device on its side and on the bottom side I placed a small desktop fan blowing air. Around 8 hours later, memtest86 finished succesfully (all 4 passes). A few hours later I restarted the test (didn't even touch the device) and it rebooted during the 4th pass.
Highest RAM temperature I've seen during this whole time was 51c/124f (temp stayed under 45c/113f most of the time). I don't think I can ask for lower temperatures during a stress test.
Then I decided to install one m2 and install windows 10 (since there is a plethora of stress/burn in tests for windows). I fully updated windows and run an old passmark burn in test utility - sometimes it finishes with no errors, other times it fails with errors and one time it rebooted during the test. Again, I barely touch/move the device, I just wait for a bit for it to cool off and restart the test. At no point does it feel uncomfortably hot and the new psu barely feels warm. note : during these tests I had a cable attached to ETH0 and a mouse along with the RAM and m2 (and in all cases/test I also have monitor and keyboard attached).
At this point I think I have to give up or buy another RAM stick. I've already contacted my supplier, will send them my current RAM stick for them to test but will probably take at least a week for them to get back to me. If they say the RAM is fine I'll probably buy a 16GB ram module that someone else suggested in this thread.
I've already ordered a NOCTUA NF-F12 INDUSTRIALPPC-2000 120MM PWM, should fit nicely on top of the device and I think I can screw it on (albeit scratching the case/heatsink) and an 8cm fan for the bottom (there seem to be screw holes for 8cm and 6cm fans). I think I can connect them directly to the motherboard but for now I'll just connect them to a separate small psu. I'm pretty sure at this point that 'fanless' is just a gimmick, I'm wondering how many people actually use these devices without a fan of some sort ...
Some notes of interest :
1. I have no motherboard drivers from CWWK, I had to download drivers for the i226 from intel and fully update windows hoping that windows update would download the appropriate drivers.
2. I cannot read CPU temperature in windows 10 (tried open hardware monitor, speccy etc). I can however read m2 temperatures and the highest it got was 62c/143f. I didn't spend much time trying to figure this out.
3. For a couple of days, every time I started/rebooted the device (either by removing power cable or ctrl-alt-del), there was a very long beep (I timed it at ~14sec, yes fourteen seconds). When the beep starts, monitor exits sleep mode but is blank, when the beep stops I see post screen and the device boots normally. What's up with that ? Can't find what the beep means in any bios error code list ... Now it does this occasionally but can't figure out why (it can happen after a failed test but can also happen after just powering on the device).
4. EDIT : forgot to mention that I tried to copy files to/from the device and only got around 550-600 mbps, it could be the drivers I found on intels site. ETH0 is connected to a 1gbps switch, every other device on that switch can do proper 1gbps apart from this one ... I didn't bother connecting it to a 2.5gbps switch yet.
As expected, no answer from CWWK. I don't know if they're a real company or just a couple of guys in a warehouse packaging/rebranding hardware they bought from somewhere else. I used the contact form provided on their webpage, is there other means of reaching them ? Perhaps a 'proper' BIOS with more options would help me sort this out.
I don't know what else to try. Anyways, I'll provide another update if I ever figure this out, thank you all for your time !
Last edited:
I have now fully loaded out my 4xM.2 board. I have a SN770 1TB, a SN560E 2TB and 2 Crucial P3 Plus 4TB drivers. I have the V1 version of the 4x2.5gbps port CWWK N305 Variant with the tall fins and lacking the USB 3 on the front. I do believe there are different. I just got a new bios from them because the one i was using first though it enabled the 4xM.2 board. disabled the wifi m.2 slot. The new bios i just got should enable the M.2 slot for the wifi as well. We will see when i get a another drive to test it. I am running unraid now with 11TB of storage.
I really like the Crucial P3 plus drives because they are running alot cooler than the WD drivers. What exact system do you have.
thanks that one does look good quality and looks like its gone through more testing to make sure its safe and passed safety checks.
My Cwwk N5105 V5edition 6 port mainboard started to develop hardware faults (1 ethernet port stopped getting detected and usb port failure) , but I spoke to cwwk on aliexpress via their chat system ( click order button then chat bubble icon or try aliexpress chat system)
And luckily I am still within 1st year warranty so they are sending out a new mainboard, not sure if its a mainboard or entire unit replacement but just glad I have not lost money on this.
Excellent support from Cwwk in regards to support and warranty assistance.
And luckily I am still within 1st year warranty so they are sending out a new mainboard, not sure if its a mainboard or entire unit replacement but just glad I have not lost money on this.
Excellent support from Cwwk in regards to support and warranty assistance.
Last edited:
Just some random info; You could use metal cable ties to strap on fans on the top or I just poked 2 small holes in my cwwk units top lid through the air vents cover and put cable ties around the fans to hold the fan to the top lid. The better noctua fans have rubber mounts so you won't get any vibration issue at least this way it just blows through the air mesh cover and onto the mainboard and hot components.
Make sure you test which orientation is best, I found laying it flat makes it hotter but on the side its more cooler but a risk it can topple so make sure its secure.
See the issues a lot of folk are getting is actually well known on this thread, these newer N series units are perhaps on rev A or B or 1 or 2 (not sure) this is why many tried to wait for the Revised versions like the popular cwwk N5105 v4 and V5 series, by the 4th and 5th revision things were more ironed out and less problematic, this may not be why you are having issues (try the different memory or simpler nvme module) etc but if you see the hellish issues from page 1 through page 30-40 its mentioned much especially on the N5xx series early on.
Cwwk is real and imo still the best supplier for support and warranty especially over kingnovy and topton who are hit and miss, maybe try using the aliexpress message service since he has a cwwk store there (see my post above), he answered my messages within 1-3 days but they are certainly more busy lately so it may take some time. If you do get contact with him ask him your questions and maybe ask if there is a newer bios update, but I do recommend if possible back up your old bios and be careful on the side.
I am confused where the mainboard drivers are for cwwk mini routers the best page I found was here so I just grabbed the intel i225/6 drivers separately from intels site and then intel chipset self checking drivers website, you can also try driver booster free edition it finds majority drivers for me most of the time.
Having said all the above, sometimes things just don't work or gel right, nothing imo beats getting a branded motherboard from asus or Asrock ie an N100 mini itx board and an intel i266 nic card and your own case+p/s unit, id bet the reliability, compatibility and overall driver/support would be the best of the best and frankly its still something I have considered.
I don't want to tempt fate, apart from my initial issue with the memory, the N100 6LAN has been fine on the stock BIOS. Temps of about 38C when relatively idle, without re-applying the thermal paste. Time will tell... I'm not going to touch anything while there are no issues!
Next unit will be for proxmox and NAS - so will be watching this thread carefully for updates.
Next unit will be for proxmox and NAS - so will be watching this thread carefully for updates.
Part of the problem with this thread is it got so confusing because there was little consistency in regards to the hardware being discussed. There is allot of discussion about devices that have nothing to do with the title.
I was trying to turn on ASPM in the BIOS for pcie but it didn't work (could not read from the disk or didn't boot). I got tired of being clueless and trying to save a watt or two. I thought I've seen a better idle power for my device before but wasn't methodical enough and did just too many changes at once. Perhaps it is just my NVMe drive and three active NICs. I'll play with the device in other ways. Bios was not fun
Hey guys,
just little update:
I have the CWWK N305 4 NIC (V2) with the x4 NVME Board - it also came with the other board that combines the NVME + WiFI key preinstalled.
I put in 2 (certainly overkill) Micron 7450 MAX 480 GB (on the x4 board) and installed Proxmox in a mirror. Also, the Wifi board hosts a Coral TPU.
The CPU, even when I benched it, never really approached 50 Celsius, and idles at about 35 (under 30 with USB fan).
I ran into in issue with the nvmes though, they got pretty hot. 80 degree or more, one even has a high temp counter in smart. So I purchased the lowest nvme coolers I could find on amazon ( M2 SSD Kühlkörper,3 Satz M.2 Wärmeleitpad,Kühler-Set Heatsink,Aluminium Cooler mit Silikon Wärmeleitpad,Halbleitergeräte Kühler,Heatsink 2280 PCIE NVME CPU-Kühlung für 2280 M.2 SSD M2 Kühlkörper : Amazon.de: Computer & Zubehör ) [german, sorry, but as a referenz], as well as a 80mm Noctua fan.
I used the onboard fan pins (with the adapter that came in the box), routed it outside the case through one of the WiFi holes (with some jumper cables), mounted the 80mm fan on the bottom, and connected it - works like a charm.
But the most important information here mby: even with these very low profile heatsinks, the bottom panel wont go on again as before (certainly not if you use the heatsink rubberbands)
BUT: There was an extra bag with screws and rubber rings in the cwwk box, which I first had no idea what to do with. Turns out, when you put the rubber rings on the cases screwholes and use the 4 screws, which are longer than the default, you can screw the bottom panel back in place and it perfectly fits with the heatsinks
It's ridiculous that these things turned out, and even that CWWK had put in these small, super useful extras.
Now the NVMEs run 20 degree cooler, and the fan barely spins.
Currently running Homeassistant and OpnSense in Proxmox, plan on adding PiHole, Frigate with the TPU and Plex and/or Unmanic with the iGPU.
just little update:
I have the CWWK N305 4 NIC (V2) with the x4 NVME Board - it also came with the other board that combines the NVME + WiFI key preinstalled.
I put in 2 (certainly overkill) Micron 7450 MAX 480 GB (on the x4 board) and installed Proxmox in a mirror. Also, the Wifi board hosts a Coral TPU.
The CPU, even when I benched it, never really approached 50 Celsius, and idles at about 35 (under 30 with USB fan).
I ran into in issue with the nvmes though, they got pretty hot. 80 degree or more, one even has a high temp counter in smart. So I purchased the lowest nvme coolers I could find on amazon ( M2 SSD Kühlkörper,3 Satz M.2 Wärmeleitpad,Kühler-Set Heatsink,Aluminium Cooler mit Silikon Wärmeleitpad,Halbleitergeräte Kühler,Heatsink 2280 PCIE NVME CPU-Kühlung für 2280 M.2 SSD M2 Kühlkörper : Amazon.de: Computer & Zubehör ) [german, sorry, but as a referenz], as well as a 80mm Noctua fan.
I used the onboard fan pins (with the adapter that came in the box), routed it outside the case through one of the WiFi holes (with some jumper cables), mounted the 80mm fan on the bottom, and connected it - works like a charm.
But the most important information here mby: even with these very low profile heatsinks, the bottom panel wont go on again as before (certainly not if you use the heatsink rubberbands)
BUT: There was an extra bag with screws and rubber rings in the cwwk box, which I first had no idea what to do with. Turns out, when you put the rubber rings on the cases screwholes and use the 4 screws, which are longer than the default, you can screw the bottom panel back in place and it perfectly fits with the heatsinks
It's ridiculous that these things turned out, and even that CWWK had put in these small, super useful extras.
Now the NVMEs run 20 degree cooler, and the fan barely spins.
Currently running Homeassistant and OpnSense in Proxmox, plan on adding PiHole, Frigate with the TPU and Plex and/or Unmanic with the iGPU.
I got the "M.2 Accelerator B+M key" for nvme and got it to show up - haven't passed it through to anything yet, but it appears for example in the OpnSense VM under Hardware -> PCI Device -> Raw Device as 'Coral TPU'
It was a bit frickly to install the proper drivers in Proxmox though, apparently something was changed and older tutorials wont work anymore.
The issue happened again today.Hi,
I've got my N100 unit a few weeks ago and have been battling with repeating issues where all interfaces would stop responding BUT seem to keep some existing connections alive. A reboot fixes the issue at first, but after some time (13h, 12h, 62h) connection is lost again.
I've looked into existing issues, but other than the issues with EEE (guru3d.com) I did not find any more reports.
Hardware
- CWWK 12TH GENERATION N SERIES 8-CORE NEW MEMBER AFFORDABLE VERSION
- 8 GB Crucial RAM (CT8G52C42S5)
- 128 GB Patriot P300 SSD
- pfSense CE 2.7.2-RELEASE
My setup consists of two pfsense boxes in two locations:
- J4125 Topton at "Main" - 192.168.128.0/24, Modem 192.168.100.1
- N100 (this one) at "House" - 192.168.4.0/24, Modem 192.168.8.1
- Linked by wireguard tunnel: 172.16.16.1 for Main, 172.16.16.10 for House. Main has a public IP
From what I was able to diagnose:
I've uploaded the system log from the last issue, the connection loss started at Jan 14 23:15:08 where my monitoring could no longer ping 8.8.8.8
- LAN side (192.168.4.1) gets unreachable (WebUI, Ping) and stops routing data to the internet
- pfSense can also not ping devices on LAN like AP on 192.168.4.5
- WAN also reports connection loss (Gateway Monitor: Modem IP 192.168.8.1)
- Wireguard Tunnel still established
- WebUI accessible, but started to degrade (Jan 15 08:27:25 in log ) until no longer accessible (Jan 15 09:57:28)
- Main Wireguard still kept showing successful handshakes until unit was physically turned off (Jan 15 19:46:27)
Gateway monitoring actions for WAN were deactivated because I suspected pfSense was killing WAN because monitoring stopped working.
pfSense System Log
pfSense System Log. GitHub Gist: instantly share code, notes, and snippets.gist.github.com
At the moment, it has been removed from active duty and moved into a lab setup to investigate this issue.
Due to the long times before issues occur, this is rather tedious to debug. So I'm looking for any ideas / help.
Steps / Investigations I've identified so far:
* Reproduce in lab (currently running)
* Disable hardware checksum offload
* Reset BIOS, Reinstall pfSense from scratch
Thanks in advance for your ideas!
Both WAN and LAN are not routing, but luckily I got in via the wireguard tunnel, even though that's running over it's WAN connection.
What caught my eye is the load average: 6.25, 6.3, 6.25
Also, System Activity looks very suspicious with two if_io_tqg_* processes right at the top.
Now, I'm unsure if this is just another symptom. As it's related to task queues of iflib, I'm leaning towards symptom instead of root cause.
Still looking for ideas, next steps will be disabling Intel SpeedStep and connecting a USB Ethernet adapter as someone else mentioned the USB-NIC keeps working even when the i226 based ones stop working.
Code:
last pid: 27087; load averages: 6.54, 6.33, 6.25 up 0+23:35:33 09:05:04
340 threads: 11 running, 307 sleeping, 22 waiting
CPU: 2.4% user, 0.1% nice, 6.4% system, 0.0% interrupt, 91.1% idle
Mem: 306M Active, 400M Inact, 739M Wired, 56K Buf, 6185M Free
ARC: 281M Total, 136M MFU, 137M MRU, 1308K Header, 6571K Other
246M Compressed, 556M Uncompressed, 2.26:1 Ratio
Swap: 1024M Total, 1024M Free
PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
0 root -60 - 0B 1648K CPU3 3 151:25 100.00% [kernel{if_io_tqg_3}]
11 root 187 ki31 0B 64K CPU0 0 23.1H 99.07% [idle{idle: cpu0}]
11 root 187 ki31 0B 64K RUN 2 22.6H 95.46% [idle{idle: cpu2}]
0 root -60 - 0B 1648K CPU1 1 151:32 94.38% [kernel{if_io_tqg_1}]
22678 unbound 68 0 36M 22M select 2 0:00 1.56% /usr/local/sbin/unbound-control -c /var/unbound/unbound.conf stats_noreset
400 root 26 0 141M 54M accept 2 0:21 0.78% php-fpm: pool nginx (php-fpm)
11 root 187 ki31 0B 64K RUN 1 20.2H 0.00% [idle{idle: cpu1}]
11 root 187 ki31 0B 64K RUN 3 20.1H 0.00% [idle{idle: cpu3}]
...Do you have photos of that? I was looking at the exact same heatsinks. If I understood correctly you said that by default the panel won't fit with the heatsink but there is a way to make it fit with rings and longer screws. Yeah, I would love to see what you did and also with the fan. I won't make you disassemble if you didn't do it tho!
I hope mine arrives today or tomorrow. Same model and I have pretty much same ideas as you.
Very hard issue to diagnose, but id go back to raw basics and raw set ups as in check cabling and maybe try another nic on the PC side or other devices, update latest bios and reset bios to factory settings and no tweaks, latest pfsense builds and keep the set up as simple as possible at first depending on your set up ie isp modem to N100 router box maybe with just 1-2 devices connected and nothing like WG/packages but just clear net running through.
And if it works fine after a day or 2, slowly re-attach other devices and rinse and repeat the steps with WG and other packages till you find out what is causing the issue.
I have sometimes suspected these i226s could be a tad flaky I recently had an issue where my intel i226 ports would just drop and 1-2 ports would work but I updated to windows 11 with latest drivers and touch wood no funny business for now, but there are still people on the net that believe intel did not really fix the whole Intel i225 let alone i226 hardware issues and some skipped it for those intel 10gb nics with backwards 5/2.5gb support which are apparently far more reliable for remaining stable connections.
Apparently the realtek 2.5gb nics have a slew of connectivity issues also.
After getting the 4-1 unit with 5 m.2 slots, it has been a disappointment and i advice people to stay clear of this company. The 4 m.2 expansion slot did not work out of the box. No problem, bios setting, read every post here. Enable it ... nothing. Moved nvme drive around, nothing. A few hours of rebooting, checking settings, absolute no life from this expansion board. The mother m.2 slot works, without issues.
Second. Jesus does that draw power! I am seeing 11W for a n100, when my cheap ass Chuwii Larkbox n100 did 3.5W using the exact same nvme drive/unraid loaded. Did not even bother trying to full power (16W = 3410 on the larkbox aka, full performance), because the heatsink was heating up way too much just from its idle. Putting another 14W into that heatsink was going to turn it not toast but hot as hell.
Cwwk support? None-existing ... So buyer beware, your wasting your money if you buy this, thinking its a nice router/nas solution. Maybe its a good router but at 11W? No, that is insane high when a basic itx/4600g runs at "only" 18W. Or my 7840HS that does 9W/11W(windows). Can just as well run a full blown PC at that point.
I have been happy with most Chinese mini-pc solutions, minus a few things but most i was able to fix. But this is the first unit that arrived semi-broken. Yay ... Now considering i had a CWWK itx board in a nas and that board was also absolute trash (20W idle for a atom) with a bios that loaded so slow, that you saw the ascii lines loading. I needed to have known better but it looked so interesting, ...
Stay away folks, build your own stuff or use some other brand their minipc and rig something up yourself.
Second. Jesus does that draw power! I am seeing 11W for a n100, when my cheap ass Chuwii Larkbox n100 did 3.5W using the exact same nvme drive/unraid loaded. Did not even bother trying to full power (16W = 3410 on the larkbox aka, full performance), because the heatsink was heating up way too much just from its idle. Putting another 14W into that heatsink was going to turn it not toast but hot as hell.
Cwwk support? None-existing ... So buyer beware, your wasting your money if you buy this, thinking its a nice router/nas solution. Maybe its a good router but at 11W? No, that is insane high when a basic itx/4600g runs at "only" 18W. Or my 7840HS that does 9W/11W(windows). Can just as well run a full blown PC at that point.
I have been happy with most Chinese mini-pc solutions, minus a few things but most i was able to fix. But this is the first unit that arrived semi-broken. Yay ... Now considering i had a CWWK itx board in a nas and that board was also absolute trash (20W idle for a atom) with a bios that loaded so slow, that you saw the ascii lines loading. I needed to have known better but it looked so interesting, ...
Stay away folks, build your own stuff or use some other brand their minipc and rig something up yourself.
Thanks for the answer.
So far, the unit has been taken out of service in the active network and placed into a lab.
There, it's WAN port is connected to my main network and it's LAN port is connected to a spare accesspoint.
Only thing I brought with me was an old patchcable which now has also been replaced.
Funnily, after switching the old patch cable for a new one, the unit also failed to reconnect.
But yeah, I suppose the next try will be installing a BIOS update and installing everything from scratch.
If I were you, I would make a bridge interface in Proxmox with two nics, assign a virtual nic and the wan nic to opnsense and call it a day. Then you can connect your pc and switch to the physical nics that are connected to that bridge. No need for new networks and works fine. Bridging does cause some cpu overhead so isn't the best option for high speeds and slow cpus but unless you're going to be saturating those 2.5G nics I wouldn't stress about it. Other option is to do it in opnsense: How to set up a LAN Bridge — OPNsense documentation
That's the thing. It's in a rather cramped place atm, sandwiched between the router and a switch, so it got a bit warmer (CPU ~40-50 degrees with average stress on OpnSense). I also use it upside down with the fan on top for obvious reasons, until I have a more permanent solution for it all
Thank you. Looks good.That's the thing. It's in a rather cramped place atm, sandwiched between the router and a switch, so it got a bit warmer (CPU ~40-50 degrees with average stress on OpnSense). I also use it upside down with the fan on top for obvious reasons, until I have a more permanent solution for it all
View attachment 33864
I received mine today. N305 from CWWK. It is gray and not black. I haven't seen the thermal paste thing yet. So far is 33C in idle and maximum I saw so far with stress-ng is 58C. It has 32gb corsair and 990 pro which seems to sit at 47 after the stress. So far so good.
I think I will leave it as is now but we will see when I add more machines and opnsense. I still need to learn more about proxmox
Last edited: