eribob

No problem! Someone with more knowledge will have to decipher the meaning of these qemu arguments though... 😃

On my 3950x GTA V runs at around 100-130FPS, Rx 580 with 4GB 1080p. I have 8 cores/16 threads to the VM instead of 4/8 as you ask for but that will hardly make a difference since single core performance is much more important for gaming.

I had a similar issue. Windows 10 VM boot-looping with blue screen. Ryzen 3950x. Solved it by adding the following to the end of the VM xml as suggested elswhere on this forum: <qemu:commandline> <qemu:arg value='-cpu'/> <qemu:arg value='host,topoext=on,invtsc=on,hv-time,hv-relaxed,hv-vapic,hv-spinlocks=0x1fff,hv-vpindex,hv-synic,hv-stimer,hv-reset,hv-frequencies,host-cache-info=on,l3-cache=off,-amd-stibp'/> </qemu:commandline>

Dear community! I am very pleased with my unraid machine and it gives me a little punch of joy in my stomach every time I open the dashboard. It is a continuous work in progress and I feel that better GPU:s are the next step for me. Currently I have an Rx 580 4Gb (2.5 slot) and a R7 370 (2 slot). It works ok but gives me some problems. 1a. The Rx 580 is passed through to a Windows 10 VM and every time I restart or shut down the VM it goes into what must be a version of the reset bug. When I try to launch the VM again I see the BIOS logo but just as Windows is about to start the monitor goes to sleep and the VM pauses. I have to force stop it and run the reset script provided by Spaceinvader One to be able to start the VM again. 1b. The card is so wide (2.5 slots) and large that it covers other PCI-E x 1 ports on my motherboard that I wants to insert USB cards into. 2. The R7 370 is ok for linux VM:s and the 2 slot size is also ok, but my Catalina VM does not accept it. Maybe this can be solved with special drivers but I would rather not go there. I am looking for 2 new GPU:s that do not need to be very powerful. I play some games but mostly CS:GO... The Rx580 4GB is adequate. So somewhere in that neighborhood or a bit more powerful is fine. Being able to buy used is only a plus. I think that new GPU:s are very over-priced in general. At least one of them needs to be AMD so that my Catalina VM will accept it. The other for the windows VM can be NVIDIA or AMD. Another requirement is that it is possible to run them smoothly without reset bugs or other issues. I want to be able to restart my VM:s! 😃 One final thing: I do not want to insert a third GPU for unraid only if not absolutely necessary. As it is today I do not need that, and my CPU and motherboard are really too limited in PCI-E express lanes. Thank you so much in advance for any advice!

I disabled the option in the Parity Check tuning plugin "pause and resume array operations if disks overheat". I had warning disk temperature at 45 and critical at 55 (I believe it is default, since I cant remember ever changing those values). I now raised the warning to 50 and critical to 60 as well. After disabling the "pause if overheat" the rebuild process has been progressing without problems (now on 39%). So most likely it was pausing due to temperatures approaching the warning level. Since I have important data on my array and no parity until the rebuild is finished, I want to await the rebuild process now. So I do not want to try and reproduce the error. Thanks again for quick support!

Genius! It actually said so in the logs, I am too stressed to check properly. Thank you!

Hi again! I am following your instructions. I replaced the disk data cable and removed it from the array. After that I re-inserted it and the disk is rebuilding. However, the rebuild process keeps getting paused with the message: Parity Tuning Operation: 2020-09-02 16:05 Notification unknown action: recon D1 (1.6% completed) Pause I can resume the process again when it pauses and it will run for another couple of minutes or so, but then the same thing happens again. The system log also mentions the drives being overheated. Is that causing the recon D1 problem? Sep 2 16:00:34 Monsterservern kernel: md: recovery thread: recon D1 ... Sep 2 16:05:01 Monsterservern parity.check.tuning.php: Paused unknown action: recon D1 (1.6% completed) : Following drives overheated: 34 34 34 31 Sep 2 16:05:01 Monsterservern kernel: mdcmd (44): nocheck PAUSE Sep 2 16:05:01 Monsterservern kernel: Sep 2 16:05:02 Monsterservern kernel: md: recovery thread: exit status: -4 Sep 2 16:08:04 Monsterservern kernel: mdcmd (45): check Resume Sep 2 16:08:04 Monsterservern kernel: md: recovery thread: recon D1 ... Sep 2 16:10:02 Monsterservern parity.check.tuning.php: Paused unknown action: recon D1 (2.1% completed) : Following drives overheated: 34 34 34 31 Sep 2 16:10:02 Monsterservern kernel: mdcmd (46): nocheck PAUSE Sep 2 16:10:02 Monsterservern kernel: Sep 2 16:10:03 Monsterservern kernel: md: recovery thread: exit status: -4 Perhaps I should remove the side panels from the case and attempt to continue? monsterservern-diagnostics-20200902-1614.zip

Ah, thank you. I will try that. /Erik

OK thanks! So replace SATA-cable and then rebuild parity?

Hi! My array has been working perfect until today. One of the disks were suddenly disabled due to read errors. The SMART report seem to indicate that the disk is healthy however? (It has FAILED: Never on all attributes as far as I can see). I have posted diagnostics - should I replace the disk or can this be some other kind of bug? I never had a warning from unraid about the disk before today, which is strange if it was failing. Best regards Erik monsterservern-diagnostics-20200902-1225.zip

Thank you for a swift reply! I will ignore it then. /Erik

Hi there! Today my server suddenly rebooted without reason and is now doing a parity check because of "unclean shutdown". Soon after I got a error message from the "Fix common problems" plugin saying that I have "Machine Check Events detected". It recommended me to install "mcelog" via the nerdpack plugin and post diagnostics here for help. So I have done this now! monsterservern-diagnostics-20200803-1006.zip Thank you for any assistance! My system spec is as follows, if you need it: Ryzen 9 3950x on Asrock x570 taichi. 80GB of Corsair vengeance LPX DDR4 3200MHz RAM.

Hi all wise people! Yesterday I installed a PCI-E USB card to pass through to a second VM. The card is this one: https://www.inateck.com/inateck-ktu3fr-4p-usb-3-0-pci-express-karte-mit-4-usb-3-0-ports.html I chose it because the FL 1100 chipset was recommended in another thread here on this forum. To fit it between my GPU:s I had to use a PCI-E x1 -> x1 riser cable from Kolink (https://www.caseking.de/kolink-pci-e-x1-auf-x1-riser-flachband-kabel-19-cm-grau-blau-zurc-005.html) The card is recognised by Unraid and shows up in the VFIO-PCI config plugin in its own IOMMU group, however, when I start the server with the card installed it behaves very strange. The cache drive starts to disconnect and reconnect with the system repeatedly, resulting in it being registered as an unassigned device, and when I try to update my VM xml I get an error message saying that the disk has entered a "read only state". I also get warnings that my cache drive is unreachable. Further another SSD that I pass through to a VM using "/dev/disk/by-it/[DRIVE ID]" is suddenly not found. Also during startup, instead of endning with the login command line prompt and the server IP address as usual, I get the following screen: It never goes further, but the server is eventually reachable over the network as usual anyway. My initial thought is that this could be a power issue, that my Corsair 750W PSU is stretched thin (i have a ryzen 3950X, 64GB DDR4, Rx580, R7 370, 4HDDs, 3 SSDs and 2 NVME drives, as well as some case fans). Do you agree or is there another possible explanation? Against this power theory, the problem occurs even though the system is not under any substantial load however. In addition, I tried to remove the SATA power cable from the USB card, but left the PCI-E connection inserted and the problem persisted. It did not go away until I disconnected the card from the PCI-E port. Another theory is that maybe my motherboard shares PCI lanes between some of the SATA ports and the PCI-E x1 port that I want to use for the USB card. I have an asrock taichi x570 board. I forgot to save the diagnostics! I can re-insert the card and recreate the error if diagnostics would be helpful though. Thanks for support in advance!

Hi! I finally built my 3950x unraid system and I am very happy with the performance increase so far! I have done some benchmarks today with my windows VM, and I would just like to ask if my numbers are to be expected or if there are tweaks to the xml that could improve performance? Precision boost overdrive is turned on in the BIOS. Cooling the cpu with noctua NH-D15. To test, I assigned all cores to one windows VM and shut down all dockers and other VM:s. It resulted in: Cinebench R20: 8600-8700 multi-core Geekbench 5 CPU: 1186 single-core, 11470 multi-core For actual use I have isolated the last 8 cpu:s (8-15, 24-31) and gave them to the VM along with 32GB of RAM. This resulted in: Cinebench R20: 4552 multi-core Bare metal performance seem to be around 9100p in CB20 and 14300 in geekbench, meaning I get around 5% decreased performance in CB20 and 20% in geekbench. Is this to be expected from the virtualization?

You mean RX 570? It seems like an NVIDIA card should work then as primary, after dumping the vbios. The 1070 is not so bad, at least it seems stronger than my Rx 580 and I do not need super GPU power... w00t? 😃 It is all coming together nicely I think! I will wait until my wallet feels full enough and then upgrade and then harass you again once everything fails...

It seems to be 2.2 slot: https://www.sapphiretech.com/en/consumer/nitro-rx-580-4g-g5#Design I will look into gigabyte options! Can you recommend a GPU that is known to function so that you can pass it trough even though it is the primary GPU in the system? I see. So the latency is there even when you run bare metal? I guess my question is if the latency on a VM in a ryzen CPU server is so much higher than the latency on a intel CPU server that intel is clearly a better option if you want the ability to game on your VM? I understand that. Is there a workaround so that you can pass through your primary GPU even though the BIOS snags it at startup? EDIT: Sorry 0.5s of googling found me the spaceinvader video... So I will try to get my hands on a decent used NVIDIA gpu then.

That is super cool to hear! Very close to what I am aming for. I have acctually considered the 3900x as a budget option. How did you manage the 2 GPU:s? Which one do you have as primary? As testdasi said, it needs to be one that can be passed through to a VM despite also being used by unraid at startup? How many CPU cores do you allocate to each VM and to the dockers? Do you game on your VM:s? In that case, does it work ok with more modern titles? The RAM is 4 x 16GB? Are there any major issues that you have had during the setup? Thanks! /Erik

I looked at a manual for the Asrock Taichi x570 board that I have been thinking about buying and this it what it says with regards to PCI-E and M.2 slots: This should mean that if I populate all of them using two GPU:s and a HBA I would get PCIE 4.0 x8 bandwidth (16 GB/s) for the GPUs and x4 speed (8 GB/s) for the HBA. The bandwidth for the GPU:s should probably be fine right? I will not be running the latest and greatest GPU:s since I am not a gamer (currently have a rx580 4GB card I bought used). 8 GB/s max for up to 8 drives that could be attached to the HBA, thats 1 GB/s per drive, which is almost 10x more than what most mechanical drives can deliver anyway, and rarely are all drives accessed at the same time right? So this means that I can not use the third M.2 slot, but I was looking for max 2 slots anyway (one for each VM is what I envision...) So the 24 lanes provided by the CPU + the extra lanes provided by the x570 chipset should be enough for everything???? This is a quite expensive board though. I will look at a cheaper option as well and come back with how that would work. Another problem is, since I want to use the 2 GPU:s for the 2 VM:s, I need to be able to pass both through even though one might be used by Unraid? But I think I have read somewhere that this can be acieved? Perhaps spaceinvader one has a tutorial...

Many thanks for the answers. Perhaps the scales are in favor of a x570 board then. I have been curious about the PCI-E 4.0 that they support. That would double the bandwidth of the only 24 PCI-E lanes that the 3950x has in theory, but I dont know if it will be true in practice... I will look at Gigabyte boards. And do modest tweaks for improved latency then. Is the latency a problem with KVM or is it specific to unraid? I expect it has to do with KVM, just wanted to ask... Also, will the upgrade to the newer kernel in 6.9 help in any way? Too bad threadrippers are so expensive hehe...

Has noone tried a similar setup akd can give me any advice? I am most interested in potential latency on VMs that I heard could be caused by a ryzen cpu. Thanks! Erik

I am planning for a server upgrade. My current build is 2 x e5 2670 and I run my main computer as a VM from them. I am really happy with the general setup bu I feel that the 2.6-3.3ghz speed is a little too slow. For example cities skylines runs at about 15fps now that my city has grown a bit. Lowering the graphics settings does not improve that, which makes me think it is a cpu issue, a lot of computations for all the inhabitants in my city perhaps? The 3950x is such a powerful chip and even though I am not editing 4K video I think this kind of build has the potential to fully utilise all the fast cores it provides! I use my computer for programming, some statistical calculations for my research, and sometimes gaming. I would like to be able to do the following: - Run 2 VM:s, with a dedicated gpu for each of them. I mainly use win 10 but want to have a Mac as well for fun - run my 20 or so docker containers - Use my SAS expansion card for additional hdd:s - Preferrably have at least 2 nvme drives. - Be able to upgrade to 128gb ram later I don’t think I need pcie gen 4 so I would like to use a much cheaper x470 board if possible (or do you see a need for x570 here?) I only need cpu motherboard and ram. The rest I will re-use. I made the following list but I would like some input from you experts as it is a big purchase! Will the board accept 32gb ram sticks for example? Do you see any other major issues here? https://pcpartpicker.com/user/eribob/saved/8y3j23 what about gaming on a vm on a ryzen server? I read somewhere here that there might be performance issues? I am no competitive gamer, just want to be able to play for fun occationally. looking forward to your insights!

Hi! I rebooted normally (just like someone else in this thread) and the start button is back. Just wanted to post my diagnostics in case someone was interested in investigating this bug. I am also impressed by the speed with which you people reply in this forum! Thanks.

Hi! Same error here. I switched out an unassigned HDD for an SSD and in that process I acidentally unplugged my parity drive. Started up the server and realised. Shut it down, plugged the parity drive back in and restarted. The parity drive is now back but the array start button is missing. See screenshot and diagnostics. monsterservern-diagnostics-20200321-1351.zip

Hi, yes i can. I can not restart the vm through windows start menu from inside the VM, but I can restart it with the unraid interface. Quite often I have to ”force stop” it and start again though. restarting the server is not necessary though.

Hi, yes, I had no problems passing through my radeon rx 580 to both a windows and a mac os VM. best if luck!