• Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by Vr2Io

  1. Haven't If 6 disks detect on BIOS AHCI ( HBA Oprom disable ), 6 shouldn't trigger the bug. I agree you can try plug-back HBA and unplug the on-board disk to trouble shoot step by step.
  2. I use Lenovo expander with different HBA without issue, you need stress test or change / replace hardware to rule out the problem. [7:0:16:0] enclosu LENOVO SAS CARD 013a - /dev/sg20
  3. Not related power or PSU issue, just BIOS issue. Someone said some motherboard if attach too much disk then it won't boot USB anymore. Does disk connect thr HBA ? If yes, pls try disable (BIOS) the OPTROM at the HBA seat slot.
  4. Good news !! Seems some post about B450 report crash but no problem on other OS, i.e @rilles . I have Gigabyte B450i ( 2400G ), but never run with Unraid, it rock stable with Windows. Later upgrade CPU also fine. I buy a used CPU in recent, Windows stress test show have memory error, this drive me nervous in few days because the CPU can't return ..... Lucky, finally rule out one of RAM module was the cause. 👏👏
  5. Yes, no problem. If onboard have a 3rd party controller also can do that, as you only need one disk.
  6. Seems to be, monitor disk response time. If your SSD was SATA, pls try found a standalone SATA controller then passthrough it to VM. But you need reinstall that Windows.
  7. Pls check does eth1 still setting as previous, as mention, eth1 shouldn't setting same subnet as eth0. Suppose remove gateway at eth1 could fix the problem. IPADDR[1]="" NETMASK[1]="" GATEWAY[1]="" What eth1 use for ?
  8. Anyway, if IP correct, it look like not Unraid issue.
  9. You can change it 1. Setting page -> CPU pinning 2. VIRTUAL MACHINES page -> left click on VM's icon, then edit.
  10. You use manual setting and eth1 in linkdown, btw nothing abnormal ( although eth1 config. in same subnet with eth0 ). Pls provide below test result ping ping traceroute dig google.com
  11. Several years ago, I use TS-851 with Unraid, but due to J1800 rather low end, so upgrade again ... again and retire for longtime. Few weeks ago, I change my last QTS to Unraid, it is TS-251+ ( J1900 ) For my understanding, the LCD are serial type, if you got the control protocol then you can control it through serial. I have same problem and this not happen on TS-851 ( different BIOS version, but still QW37AR3X ), I update TS-251 BIOS ** directly under Unraid ** from QW37AR32 to QW37AR36 then fixed this problem. https://forum.qnap.net.pl/download/ts-x51
  12. It look like relate idle problem, if everytime call-trace have " RIP: ... cpuidle_enter_state ...... ", then it should relate this. My suggestion was try without start array or mount any storage, let system in most idle status, then check does trace will happen within 1hrs.
  13. Change to X299 platform some month, but not all motherboard would utilize 44pcie lane and not easy to found. Cooling will be serious issue if all 7 slot with high power device. A used 7960x will available in few days.
  14. In fact nothing special need to set, passthrough NVMe, always switch between VM and baremetal, no issue in different platform. ( Pls dont mount or write it )
  15. The PCIe AER error come from on-board device, (00:01.1 bridge), not come from add-on display card. I never see AER come from on-board device, pls try remove extra setting in "syslinux.cfg" and remove "PCIe ACS override", if AER still spamming, it could be motherboard or CPU problem. Any history/background can provide, does it work normal previously ? Feb 21 21:42:19 Tower kernel: pcieport 0000:00:01.1: AER: Corrected error received: 0000:00:00.0 00:01.1/00.0 USB controller: Advanced Micro Devices, Inc. [AMD] X399 Series Chipset USB 3.1 xHCI Controller (rev 02) 0
  16. For IOMMU issue, Pls try 2 things 1. https://bytee.net/blog/amd-threadripper-kvm-windows-gpu-passthrough - amd_iommu=on - enable Windows 10 WHQL Support 2. If not work, then try disable CPU core not more then 16 ( 32 thread ) Pls also provide "lspci -P" result.
  17. Recommended setup was use UD plugin to mount a SSD which host VM / Docker ( avoid use array or share-cache ) this should solve most problem. Array not suitable for heay multiple read and write application.
  18. 1. Pls check does it detect at Unraid system device. 2. Does HBA heatsink hot ? If not, it should be dead. 3. Pls try tape B5 B6 at HBA pcie contact finger.
  19. So cheap. DIY 👍 Connect with one or two QSFP cables ??
  20. Should be two different issue SVM and IOMMU issue : Feb 21 04:44:53 Tower kernel: AMD-Vi: Can not reserve memory region 0-80000 for mmio Feb 21 04:44:53 Tower kernel: AMD-Vi: This is a BIOS bug. Please contact your hardware vendor PCIe error report: Check which add-on card cause error on PCIe bus by "lspci -P" 00:01.1 PCI bridge [0604]: Advanced Micro Devices, Inc. [AMD] Family 17h (Models 00h-0fh) PCIe GPP Bridge [1022:1453] Kernel driver in use: pcieport