MowMdown

It is a point release, edge cases like yours might slip through the cracks... The solution is a easy fix. My advice is wait for the X.X.1 update if you really want to be safe.

Something to keep in mind, if you dont have active cooling on your HBA it could also be overheating, these things are meant for a server chassis with high airflow and if you are running one of these in a consumer case, it will need a dedicated fan for itself.

Sounds like a cabling issue honestly. Bunch of your disks have UDMA CRC errors which indicates a faulty connection between he disk and the CPU. Double check all the cables are connected properly. Please reboot post new diagnostics. the syslog is 0bytes and the previous syslog is full of disk read errors on disk11 (presumably the old one)

worst case is you delete the network.cfg and network-extra.cfg from the USB and reboot unraid

going to need to connect a monitor and keyboard to the server and log in and get the IP address. either your IP address changed or you set a static ip on the server and it's no longer on the same subnet.

Please post diagnostics

One other thing, you also have appdata located on disk18, you might want to address this since appdata is set to cache preferred

See the link JorgeB posted, I would like to rule out PCIe communication errors. Also if you could reboot and post new diags that would be helpful, there is a lot of log spam and what I am looking for I cant find due to it being too far beyond what is logged.

your cache is full or at least is being reported as no free space left, it's probably crashing due to this Apr 16 00:08:25 MEDIA-SERVER kernel: pcieport 0000:00:1a.0: AER: Correctable error message received from 0000:02:00.0 Apr 16 00:08:25 MEDIA-SERVER kernel: nvme 0000:02:00.0: PCIe Bus Error: severity=Correctable, type=Physical Layer, (Receiver ID) Apr 16 00:08:25 MEDIA-SERVER kernel: nvme 0000:02:00.0: device [15b7:5017] error status/mask=00000001/0000e000 Apr 16 00:08:25 MEDIA-SERVER kernel: nvme 0000:02:00.0: [ 0] RxErr (First) Apr 16 00:08:29 MEDIA-SERVER shfs: share cache fullIm also not sure what these nvme messages are maybe @JorgeB can tell us.

please post diagnostics

please post diagnostics

FYI it's only ever been --runtime=nvidia in extra parameters. Not sure where people started getting --gpus all from...

yes Im using one right now. Also I paid $25 for 5 of these suckers.... $30 for a single one is a rip Note that it will also count towards your license storage limit

The TL;DR Is if you're on an older nvidia GPU, pascal or older, you are limited to the 580 driver branch due to nvidia dropping support for older GPUs on 590+ The 590 driver branch only supports ada series GPUs and newer. (And their workstation counterparts of course)

Post the screenshots of your docker container template pages where you configured what you did

Best thing you can do is spin up a trial and take it for a spin.

Proxmox isn’t NAS software it’s a hypervisor which is meant for running virtualized OSes. TrueNAS runs straight zfs pools so you need all same sized disks and expansion is not as simple as adding another drive. It requires careful planning of disks and pools. You’re going to get biased answers here cause we all use unraid for some reason or another. The only way to really find out if something else is better for you is to try it.

Please post your diagnostics

It’s normal that containers networked through other containers only rebuild when restarting the parent container from the docker page. There is a docker container called “ContainerNetwork-AutoFix” that will automatically repair these.

You would be better served using Tailscale plugin to securely access your server instead of exposing the unraid webgui to the internet though a CF tunnel. Using a CF Tunnel is a HUGE security risk. Tailscale is a secure private "tunnel" that only you can access.

Yes as long as you only have Parity1 and no Parity2 (parity2 the disk order matters) 1. Stop the array. 2. Tools > New Config (perverse all slots) 3. On the MAIN page, move disk6 slot assignment to the disk3 slot 4. Check "parity is valid" and start the array.

Yes, that is the correct assessment, if you are coming in fresh, you dont need the pgvector and just stick with vectorchord or whatever their docs say to use.

make sure in windows explorer you're using the IP ADDRESS not "tower.local" or whatever your mDNS hostname is.

Great news! Vlad released a patch that allows filerun to work on the iOS Nextcloud APP!!!!!! 26.2.0 is released: https://filerun.com/index.php/changelog?v=2026.2.0

if youre on the newest 2026.1.0 those var's aren't needed anymore This is all you need on 2026.1.0 version (you will need to chown nobody:users from the unraid console on your appdata folder if it's not already)