TPNuts

Nevermind unraid did it's thing and went into hyperdrive. 400gb transfer finished in 8 hours.

the drive is a SN750 which has a write speed of 2600 MB/s and I am writing at 1 MB. smh

I decided i wanted to move my old cache drive to a newer one, so I moved everything over to the array and then unassigned the old nvme and assigned the new nvme to cache pool. I then moved the transfer for appdata and system back to cache (array to cache) and started mover. The issue (or I'm missing something) with transferring my data back from the array to the cache drive using mover. is that it is transferring from 1 single disk to the NVME drive at roughly 1 MB a second. I am doing something wrong here, or is it just a large volume of docker files that is making it so slow? I feel this is going to take so long that I will be dead before it transfers at this rate. Attached is the diagnostic and a photo servo-diagnostics-20250404-1659.zip

I cannot believe we are still on this issue after 3 years. Has no one on the team resolved this yet? I have had the same problem for the last 2 years and resetting the process is only a temporary fix. It's a bandaid on a bullet wound philosophy here. Also increasing the memory size is not a solution since it will just outgrow the new limit. "nchan: Out of shared memory while allocating channel /cpuload. Increase nchan_max_reserved_memory. Sep 12 04:40:10 Servo nginx: 2024/09/12 04:40:10 [error] 2372#2372: *1895414 nchan: error publishing message (HTTP status code 507), client: unix:, server: , request: "POST /pub/cpuload?buffer_length=1 HTTP/1.1", host: "localhost"

Having the same issue as well here, it is truly annoying as it randomly works but when it doesnt work its persistent.

Anyone know how to verify the IP address of the docker? Curl command is not working on the console which is odd.

What was your performance increase just out of curiosity, I am trying to configure a win11 vm (unraid 6.10.0 rc2) with passthrough of my 3090.

So a followup to this: I essentially got up and running although 6.10.0 rc2 is not very friendly with usb passthrough on items I have. It Turns out the vbios isnt even needed and that the stubbing works well for the 3090. The only major issue is the usb passthrough as Windows 11 hates recognizing pasthrough cards and even usb items added individually. I dont know if I am doing this correctly but the drivers fo not recognize half the items. More to follow. Make, break, and recreate.

Im in the same scenerio with a 3090 Founders. It seems some have been able to leave the vbios blank but my setup doesnt post without it. When I also run the script I get either an error telling me to bind the item (even though vfio binding is already in place) or more likely it posts some file (or some random file) that is under 70kb. i have looked over the script and I think (and i may be very wrong) that the script is no longer able to pull vbios using the temp vm created. I observed that it doesnt post 1. at all. or 2. if it does it errors out. I am kind of stuck currently in limbo (or shutting the server down to go baremetal.

It was a Quarantine christmas (wife whos a physician got covid on day 1 of christmas break 🤷‍♂️) so I had a ton of time to troubleshoot, and I can pretty much confirm that the vbios dump script is borked with the new OS updates. I tried 6.9.2 and Next branch 6.10.0 RC2 trying to dump a founders edition 3090 Nvidia card and doing a passthrough. I am getting vbios sizes in 70kb range which is not right at all. I have updated the BIOS on the main board (TRX40 Auros Master with a 3970x) and still no luck. I have narrowed the issue directly to the vbios itself as I cannot get anything beyond a black screen and the log goes nuts and errors vfio on the binding to the nvidia card when the vm is run (subsequently filling the entire log storage to 100% without fail every time. I am going to post the diagnostics, logs, and xml below. Also I will post the errors asscoiated with bios dump from the script used with SpaceInvader (note I have tried enabling and disabling ReBAR and that didnt help. Some have had luck without having to dump the vbios but in my case it isn't posting without it. Any help would be appreciated. Also feel free to openly school me, I am always open to learn how to correct my mistakes. Any Help is appreciated XML: syslog: VFIO-pci: SpaceInvader vbios log:

Is it even worth changing hidden state. Didn't nvidia remove KVM checks on the latest driver?