Jump to content


  • Content Count

  • Joined

  • Last visited

Community Reputation

56 Good

About Jerky_san


  • Gender

Recent Profile Visitors

1443 profile views
  1. Advanced -> CBS -> DF there is a Memory model item that has choices like auto | distribute | channel. Channel should be selected. Looking at the numa stuff but I'll give you some updated stuff tmrw.
  2. You could boot in gui mode which will give you access to the gui of unraid where then you could assign the interface potentially and set the IP/DHCP. Though if they are unclaimed I'm unsure if for some reason they aren't being recognized but should since they are intel.
  3. Just want to confirm. In the pic of the docker you have the extra parameter but did you add that via advanced at the top right or did you add it as a variable?
  4. @commotion I should also mention that there is "another way" that kind of bypasses a lot of problems with QEMU but it involves virtualization of Unraid and running the free version of ESXi. I did that and it's basically the best of both worlds to me. It solved all my "AMD" problems since QEMU hasn't caught up yet, took away the hassle of customizing the XML heavily to fit threadripper, and also gives me the ability to better control everything. Basically unraid is now storage/dockers. ESXi is everything else like gaming machine, pfsense, and other things I've been toying with. The ONLY major problem I had was sorting out which USB controller would pass through. You could easily bypass that problem since you have 2080ti's that have onboard USB3.0's I believe. Anyways it solved all my hitching and stuff I'd have with my gaming VM in the old days. The only problem is that I sometimes get high ready time if I'm doing a heavy amount of downloading(writing/reading) and utilizing my machine at the same time. Slight stutter here and there. Also I never was able to pass the audio card so I just use USB audio that I have from my old "all in one zfs" box. Edit: ESXi 7.0 that is.. it also allows you to boot from USB which solves a lot of the old problems and slow boot times.
  5. @commotion So some other things you need to do.. In the bios under I believe at least in the zenith was "CBS" there is a place to set how the memory channels are set. You need to set it to die I believe. Look for IOMMU and make sure its set to "enable" instead of "auto". Also I don't remember the IOMMU groupings being bad on my tachi so I'm surprised they'd be bad on yours but I only had the tachi for a short time. There is some numatune settings to your xml file you can do to optimize your settings once we get performance slightly better. The reason you don't want to do them right now is because it makes the gui editor not work as well.
  6. I looked through your XML and it appears "normal/good". Before the newer versions you had to do a lot more customizations. The C-State fix shouldn't be required btw. If your getting freezing you'll need to do the option where it says "power supply idle" - normal instead of "low power". I'm unsure what this option is called in your particular board. Do you have "Tips and Tweaks" plugin installed? If you do make sure the CPU Scaling Governor is on "On Demand" and the Enable Intel Turbo/AMD Performance Boost setting is enabled as well. (Just working through the beginning things btw since I saw you watched space invader's very helpful vids. Just want to cover all the basis). On the windows install/vm. Have you tried just passing the ssd and nothing else to see where performance is landing you?
  7. I'm on my phone right now but I'll look over this next time I get to my desk. Looking at the picture you posted it appears your using a 2990wx correct? It can be slightly difficult to say the least. Edit saw the parts picker thing. So yeah 2990wx. It only has two numa so it will make it all fun.
  8. Hmm just from my experience usually balloon errors are from not enough ram allocation. Can you post the top part of your XML for this machine? There is a linux machine in your diagnostic but I didn't know if that was the right one. The diagnostic you posted seems like the machine your running on has uber amounts of ram if I read that right.
  9. If we can let's start with the obvious stuff that I gotta ask(sorry if you have done any of this) honestly don't know the exact location in the diagnostic hence the questions. Have you done CPU isolation in settings, CPU pinning? Does it stutter badly or just frame drops? Are you using GSYNC?
  10. I've tried setting it to a min I feel comfortable with but do you happen to know what happens if it does try to write and it can't fit it? Does it retry or something or can it be catastrophic?
  11. Is there a document or something that talks about things like this? Until yesterday I had no idea Unraid can switch between reconstruct write and read/modify/write mode without me explicitly telling it to. Do you know if there are other settings that can turn on/off when the system deems it required? Anyways would be nice to know of anything else so I can remember it if I'm ever trying to debug an issue.
  12. Just a quick question.. are your drives mounted via the Fujitsu raid card or via onboard? Also is there some sort of expander or something?
  13. You might could try this by I'd caution to wait for others to chime in and maybe they know something else to try. Click the little angled arrow at the top if for some reason you can't click the actual link. This may help you.. I'm unsure.. As said I'd wait and see if others know more.
  14. Sigh, I feel like I am out of my depth on this one. The network graph seems to show good performance there though I would kind of expect close to line speed. But you also said using the mover causes it to write to the array very rapidly. I am hoping someone else will chime in that can better help you. I can say looking at your sys log it's jammed packed full of these errors. It's interesting because I don't see any drives labeled disk29 or disk0. Jul 6 02:49:28 Jarvis kernel: md: do_drive_cmd: disk29: ATA_OP e0 ioctl error: -5 Jul 6 02:49:29 Jarvis kernel: mdcmd (5077): spindown 0 Jul 6 02:49:29 Jarvis emhttpd: error: mdcmd, 2723: Input/output error (5): write Jul 6 02:49:29 Jarvis kernel: md: do_drive_cmd: disk0: ATA_OP e0 ioctl error: -5 Jul 6 02:49:29 Jarvis kernel: mdcmd (5078): spindown 29 Jul 6 02:49:29 Jarvis emhttpd: error: mdcmd, 2723: Input/output error (5): write Jul 6 02:49:29 Jarvis kernel: md: do_drive_cmd: disk29: ATA_OP e0 ioctl error: -5