tjb_altf4

A word of caution... Nested virtualization using a (virtualised) windows host is still not supported for current architecture AMD CPUs, this is a Microsoft limitation. You will likely brick your windows VM trying, so make sure you backup your VM if you want to give this a go.

+1 Have been running 2x SanDisk Cruzer Fit Ultra 16GB USB3.0 for +3 years in seperate Unraid machines without issue. Strange that the USB3.1 versions are creating issues.

Unraid currently supports single cache made of 1 or many disks. Next release is meant to have multiple cache pools, which should fit your use case. You could also look at unassigned devices plugin for adding disks that are not part of cache pool or array.

depends_on is a "docker compose" flag and not a "docker" flag in itself, where unraid doesn't use docker compose natively

Try under advanced setup. I'm not familiar with netgear routers, so cannot be certain.

FYI the problem is the lack of graphics drivers, but is also impacted by certain cards behaviour with the default display driver. Tested with a large 4K screen (possibly a contributing factor). My experience: older amd gpu: garbage experience with microtext on a tiny resolution new nvidia gpu: usable resolution (but not full) with expected text size, OK experience. new nvidia gpu + nvidia unraid build (aka having gpu drivers): full resolution, expected text size, great experience I would expect new AMD with drivers to also be great, but new AMD with default driver might be a crappshoot.

The error is related to a security setting on your router, try and see if your router has this setting available and uncheck it. Note this is from a different model netgear router that may or may not be the same. Advanced -> DHCP/DNS

Try changing the install media from usb to sata, or vice versa

Same issue here on another BSD distro, solution posted in there

I gave an install a go and saw the same issues as you had using freenas 11.3 and bsd template. Similar hang is also described here: https://redmine.ixsystems.com/issues/77718 So it is some combination of vm's uefi settings or version causing issues by the look of it

This can be defined in the Settings > Docker in the GUI as long as the NICs are visible (and enabled) in Unraid. note: the Docker service must be disabled to make changes and Advanced View (top right) toggled on.

They have commented many times in this very thread

Try to avoid simulated sine wave UPS as a general rule for modern pcs, the power supplies dont like it, but CyberPower still make good units IMO.

I would avoid naming (including vdisk folder path) that uses special characters like an exclamation mark. I know this can cause issues with bash, so might be the problem here.

Do you have a specific reason for staying on beta version, and not simply rolling back to latest stable version?

This link covers the issue as being a Radarr/Sonarr one https://github.com/Radarr/Radarr/issues/3256 A quick read suggests it is limited to magnet links, try avoiding in the short term if possible, looks like a fix is in the pipeline.

I think its disabled by default these days, but you can enable it here: Scheduler > Mover Settings > Mover Logging > Enabled

Its the same file. One is on the physical disk (disk1), the other is on the user share VMs, which includes the physical disk1.

You have to use the IP that corresponds to the 10GbE connection instead of the 1GbE connection, there is no prioritising as such.

I got a pair of Dell H310s (9211-8i) off the art of server on ebay, he will flash to IT mode as part of his service.

It used to be there, find it here now: Settings > Management Access > Local TLD

Does this upgrade process only work on the default raid1 pool? i.e. will it work on a raid0 pool?

Nested virtualisation is what you want to investigate, Unraid supports it, but the VM's OS (in your case a MacOS) needs to support it also. On windows, Microsoft only supports Intel cpus for nested virtualization, even though AMD hardware can support it.

Probably doesn't need the array to be stopped, but there would be many services that need to be stopped and restarted (to push/apply changes) that it's easier to use the existing command stack that is part of array restart procedure.