Vr2Io

Members
  • Posts

    3025
  • Joined

  • Days Won

    3

Everything posted by Vr2Io

  1. First ver. actual throughput will be 300MB/s and I am not sure Unraid support newer version or not even you have 3.2 gen 2 controller & device.
  2. Like others mentions, raid controller may be the issue and I also agree that. Seems you only have 3 data disk, could you try connect disks to onboard SATA to bypass the RAID controller for troubleshoot ? I use Unarid several yrs ( up to 6.10 didn't try 6.11 ), spindown usually work well with onboard SATA or LSI HBA. Below is one of 7x24 Unraid, basically it only spinup once per day. ( Other spinup just because I access it )
  3. You should mount those new disk i.e. UD then copy in disk-to-disk in same time, after all complete, then new config the array with new disk and rebuild parity.
  4. This will disable "turbo write", keep writing one file only. Won't that low for mid-high grade NVMe. Also apply to Unraid if you use software RAID by pool / UD. This won't happen, array throughput won't over single disk throughput, so for spinner disk usually in 140MB/s - 90MB/s. If you have enough RAM cache, then network transfer to RAM will saturate the 10GNIC ( 1GB/s ) and it will "slowly" writing to array.
  5. You should try enable/disable "PCIe ACS override" or ACS in BIOS. BIOS ver. usually haven't much different.
  6. I didn't try that, I passthrough NVMe ( PCIe device ) in device page in past. As my application not performance demands, so I haven't deeply test / compare.
  7. Passthrough whole NVMe instead vdisk will got better performance. You can directly boot the NVMe in VM.
  8. The Governor ( dynamic clock multiplier ratio is different for "non-K" & "K" ) https://www.intel.com/content/www/us/en/products/compare.html?productIds=191792,186604 For example 9700 vs 9700K, clock range are 30-47 vs 36-49, so it's scale is different. In overall, 9700 will consume less power but also a bit slow. But you can config 9700K ( some BIOS ) in same scale too, so theoretical it can consume same power as 9700. That's flexibility. As different CPU almost use same amount power when idle, so "K" type CPU in normal build also acceptable.
  9. It likely USB wakeup hsve turn on, try disable it in BIOS.
  10. No matter MACVLAN or IPVLAN, DHCP always useless because docker won't get IP from router/external DHCP. This also won't happen, you should setting an IP range for DHCP purpose, and all non-DHCP application use out of this range. Some SMART enough router also work well if you bind MAC address out of DHCP range IP. You can use br'x' network mode then static a IP for each docker, only some docker must use host mode. ( Of course all with same MAC address )
  11. Yes, but those adapter was insert in vertical, it may too height. Below are more common form factor. Anyway, pls try unplug the M2 to confirm LSI HBA will work first.
  12. Just double confirm, does speedtest was desktop (ethernet) to Internet server ?
  13. The manual have mention that "PCIe 3.0 x16 slot (PCI_E3), supports x4 speed" won't available in some situation. But it is not clear if only two M2 case. When installing devices in M.2_2, PCI_E2 & PCI_E3 slots at the same time, PCI_E3 slot will be unavailable, and M2_2 slot only supports PCIe x2. Another solution was don't use M2_2, and move a M2 to PCIe 1x slot, then make PCI_E3 available. Drawback was bottleneck at that M2, but it will be the cheapest solution.
  14. Does other ethernet device also 2.5/5G capable ? If your speed test was under WiFi which provide by BGW320 then you will got over 1G speed. At BGW320, the blue port mark as 5G, does it correctly negotiate at 2.5G with your 2.5G switch ?
  15. I always reboot HA (or other VM) if I delete/clear database and no trouble. But I do in manual, shutdown in HA, then restart in VM. My VMs most are qcow2 because thin image saved lot of space. Not much idea on this case.
  16. This abnormal, are you ping from Mac 192.168.10.2 to Unraid 192.168.10.1 ? How about if at Unraid ping to 192.168.10.2 ? Seems you use two 10G port for eth0 eth1, pls ensure the port with mac address ending in E5 connect to router and E6 was connect to Mac. Then post network setting / state screen dump at Unraid and Mac. Unraid internet access work ? ping 8.8.8.8 work ? ping 192.168.1.1 work ? Mac internet access work ?
  17. TimeMaschine... or TimeMachine... Pls also try ping test to 192.168.10.1
  18. XML won't change from qcow2 to raw if VM reboot, if that something abnormal.
  19. Subnet mask wrongly set to 255.255.0.0 (/16), correct it to 255.255.255.0 In simple, use subnet mask 255.255.255.0 ( /24 ) in all setting.
  20. 1. WiFi was use to access Internet, 192.168.1.x with gateway 192.168.1.1, setting as usual at both side and always enable. 2. 10G subnet 192.168.10.x won't have gateway at both side, just blank the gateway, only use for high speed file transfer.
  21. Native Unraid haven't router / bridge two physical network port function, a common method was make a direct 10G connection between NAS ( one 10G port be eth1 ) and PC for file transfer, setting them in different IP/Subnet in static IP. Then onboard 1G NIC in NAS ( eth0 ) and PC WiFi, for access internet and manage Unraid.
  22. Seems you transfer those files in parallel, you should set it in one by one.
  23. Pls don't misleading, I am talking about false negative. Don't waste both time. And I haven't false negative issue on static file by FIP or Bunker.