sjameson

Made a few more new cables and tried them out. Same issues. Except now the speed drops if the VM is running, not just when a download is going. I have no idea how to proceed. Nothing has changed software-wise since before the problems started. Cant even download Docker container updates without the whole server bogging down to slower than a crawl. direwolf-diagnostics-20251212-2117.zip

Tried that out and didn’t have any luck… Not quite sure what’s going on… In the VM, the card says 100Mbps, but the br0 that was set up in the configuration of the VM shows 1000Mbps (in terms of network card properties). Speedtest.net gives me 8Mbps down and ~50Mbps up. But my librespeed container shows 2,500Mbps….

Didn't seem to work.... Making sure I did it correctly, i switched my main NIC in network settings to have a different one in eth0, then set the separate NIC as eth2 (my onboard NIC has dual ports and the second one took up eth1). Then i changed the network interface in my VM configuration to use eth2 instead of br0. Tried it the other way around too with my separate NIC for my Unraid side of things and the onboard NIC for my VM. I can't seem to figure this one out.

Tried swapping in a NIC instead of my onboard one and the problems are persisting. What I figured out is that the internet issues are only in my VM, not the rest of my Unraid setup. But, if I download anything inside the VM, it slows down everything (even my Docker Containers). But, the Docker containers can handle any type of bandwidth without slowdown as long as the VM is not using any bandwidth. Anyone have any ideas?

download speeds are in the gutter.... I'm thinking I need a separate Intel NIC. Can someone help me confirm? Tried a different CAT5 cable, tried bypassing my 5-port switch as well. Was going to try updating my BIOS, but won't have time for a while to get to it. Just ordered two cheap NIC's to try out, but are there other suggestions for reliable NIC's these days? direwolf-diagnostics-20251104-1936.zip

Yeah, about to be moving, so once I get everything set back up at the new place (set up and stable), I'll check for an update. The port that's throwing the error is not a slot that I can just swap from (I don't think)... I traced it all the way through and all of the components downstream are built in to the motherboard (not external devices seated in a PCIe slot). Here is the result of lspci -tv +-[0000:20]-+-00.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse Root Complex +-00.2 Advanced Micro Devices, Inc. [AMD] Milan IOMMU +-01.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge +-01.1-[21-2c]----00.0-[22-2c]--+-01.0-[23]----00.0 ASMedia Technology Inc. ASM3242 USB 3.2 Host Controller +-02.0-[24-25]--+-00.0 Intel Corporation Ethernet Controller X550 \-00.1 Intel Corporation Ethernet Controller X550 +-03.0-[26]----00.0 Intel Corporation Wi-Fi 6E(802.11ax) AX210/AX1675* 2x2 [Typhoon Peak] +-04.0-[27]----00.0 ASMedia Technology Inc. ASM1061/ASM1062 Serial ATA Controller +-05.0-[28]----00.0 ASMedia Technology Inc. ASM1061/ASM1062 Serial ATA Controller +-06.0-[29-2a]----00.0-[2a]----00.0 ASPEED Technology, Inc. ASPEED Graphics Family +-08.0-[2b]--+-00.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse Reserved SPP +-00.1 Advanced Micro Devices, Inc. [AMD] Matisse USB 3.0 Host Controller \-00.3 Advanced Micro Devices, Inc. [AMD] Matisse USB 3.0 Host Controller \-0a.0-[2c]----00.0 Advanced Micro Devices, Inc. [AMD] FCH SATA Controller [AHCI mode] +-01.2-[2d]----00.0 Phison Electronics Corporation E18 PCIe4 NVMe Controller +-01.3-[2e]----00.0 Phison Electronics Corporation E18 PCIe4 NVMe Controller +-02.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge +-03.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge +-04.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge +-05.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge +-07.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge +-07.1-[2f]----00.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Function +-08.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse PCIe Dummy Host Bridge \-08.1-[30]--+-00.0 Advanced Micro Devices, Inc. [AMD] Starship/Matisse Reserved SPP +-00.1 Advanced Micro Devices, Inc. [AMD] Starship/Matisse Cryptographic Coprocessor PSPCPP +-00.3 Advanced Micro Devices, Inc. [AMD] Starship USB 3.0 Host Controller \-00.4 Advanced Micro Devices, Inc. [AMD] Starship/Matisse HD Audio Controller

Errors are still reporting, but the bottlenecking isn't happening anymore! Not sure if it was just the error suppression or the combination of that with the ASPM getting turned off, but all good now. With the errors still being there, is it something to be worried about, or is it just a hardware quirk for ASUS motherboards?

Unfortunately, that didn’t work…. Ran cat /proc/codling to confirm it was in there too. Unless I did something wrong (honestly thats the first time I’ve edited my config).

I'm just trying to figure out what to fix... I'm hoping it isn't a motherboard issue. I've re-seated PCI slots, checked my power connections, and I'm still having issues. I think the NIC, GPU, and a few other things share the root port that is throwing errors, so the errors (which are all being "corrected by hardware") are throttling everything with interrupts. Not sure what to do besides replacing the motherboard...

My log has been filling up with errors, and it seems to be stemming from my root PCIe port (my NIC is downstream on this one). So, when i download anything on my Windows 11 VM, everything bogs down when I'm connected remotely. Unusably slow... Problems just started in the past week or so (hard to pinpoint as things seem fine when there isn't something downloading). Trying to figure out if it's chip noise, needing to update the BIOS, failing PSU, failing motherboard, etc.). Any help would be awesome! Not sure if the problems were there before or not (again, because things seem fine when there isn't a download going), but I did just update Unraid about 6 days ago. Didn't notice anything before that. Currently have Unraid 7.1.4 installed. direwolf-diagnostics-20250907-0830.zip

Got home from work today and my wife mentioned that the server was off and she had to turn it on, but it never fully came online. It was stuck on a 04 code on my ASUS Pro WS WRX80E-SAGE SE WIFI II. This is a “PCH initialization before microcode loading” step for that board. I had to turn it off, unplug the power cable, and hold down the power button for a while to drain residual power, but it worked and the server came online like nothing ever happened (besides a parity check due to an unclean shutdown). Not sure where to start on the troubleshooting side of things. I have a decent amount of drives, but my power supply is 1300W and I don’t think those would be maxing it out (just replaced two drives that had smart errors which are connected in unassigned disks so I can run extended smart tests on them). With those two new drives, I added an LSI-9300 8i HBA with the stock cables. Any ideas or help would be great! Unraid version 7.01direwolf-diagnostics-20250513-1844.zip

Just for anyone else who finds this thread down the line: SeaBIOS will not work with my setup due to the GPU requiring UEFI for passthrough. I was able to restore to a previous backup with the following steps (the custom XML VM provided above did not work as it wouldn't boot with my GPU due to the BIOS being changed to SeaBIOS): Created new nvram folder in etc/libvirt/qemu for my OVMF VARS files Copied backup .fd into the newly created nvram folder (make sure to remove the timestamp from the beginning of any backup files) Copied backup vdisk1 into my domains folder associated with this VM Copied old backup libvirt.img (xml and vdisk 1 were from late July, and the libvirt image was from back in January) and replaced the current libvirt image in /mnt/user/system/libvirt/libvirt.img Used the xml to create new windows 11 VM via xml mode Everything booted up at that point and I only lost a month's worth of software that I had setup (no important documents, just the software packages themselves). Probably going to rethink my backup strategy for VM's now and make things more automatic. I stopped doing automatic backups because I had to shut down the VM in order to do so. Not sure if I want to redo everything and try to start using snapshots instead, but I'm not sure how straightforward that conversion would be. Likely going to just do more frequent manual backups after major changes to software/settings that I want to persist. @bmartino1 I appreciate the help! Your solution will likely help a lot of other people, but didn't quite help me due to requiring UEFI boot for my GPU passthrough. Long story short: if you have a .fd VARS file for a backup restore, but your nvram folder is gone, just create a new nvram folder where it should go (/etc/libvirt/qemu/nvram/) and copy your .fd file in there and start up that VM!

I can't do anything remotely with that VM right now unfortunately, so I can't see what's going on in that machine right now. But, it looks like it booted. Just need to check on my monitor what's going on once I get home. There is still no nvram folder. Do I not need that .fd file for everything to work correctly? I see the OVMF lines are removed from the code you provided, does that not matter?

Backups done via VM Backup plugin (not as regularyl as I should've been doing, but most important files are on server shares - just software that will need to be re-downloaded). I have a libvirt file from months prior to the vdisk backup (for some reason it didn't backup the libvirt at that time).

direwolf-diagnostics-20240826-1246.zip Apologies, working off my phone today with my VM being down and didn't grab those. See attached. Stable release of Unraid. 6.12.10