itimpi

What is wrong with using the Prefer option with a suitable value for Minimum Free Space? That automatically overflows to the array when the free space on the cache drive drops below the Minimum Free Space setting so what is different that you are looking for?

The Linux 5.x kernels are meant to have a lot of new features. I wonder if this has increased the overhead?

This is the default as most people want to keep related files together.

There should be a virtio iso image file available which has the drivers to support virtio networking (the most efficient). Alternatively you can change the network card type in the VM definition to be one that Windows already has drivers for. You could also use the virtio image file to install virtio disk drivers if you want to use those instead of SATA.

In principle it should just be 'plug-and-go'. It would not hurt to have a screenshot of your current drive assignments just in case. Since you are not doing hardware pass-through the VMs should also be easy and vdisks go across without issues. The one thing that might need attention is if you have any USB devices assigned to any VM. In such a case there is a chance the ID changes and you need to redo the assignment.

Did you make sure that you used one of the Slave access modes when setting up the mapping? If not that might explain why you had to stop/restart docker. It is always recommended to use a Slave mode when mapping to an UD path.

Have you made sure when running as host that they are not trying to use the same port?

you get rid of dockers runnings by disabling the docker service. Plugins are automatically not run if starting in Safe Mode.

If any plugin in UnRAID is deemed to be suspicious you will quickly hear about it The Unassigned Devices plugin definitely falls into the category of being a trusted plugin - so much so that there is talk about its functionality being included into the core release at some point. In UnRAID there is another plugin (Community Applications) that adds the equivalent to an App Store to `unRAID (and adds an Apps tab to the GUI). That only allows plugins that have had at least a cursory check made on them to be installed. there is also the fact that as far as I am aware all UnRAID plugins are made available on an Open Source basis so can be checked by any party interested in investing the time to check this. One of the strengths of `unRAID is that there is a significant community who add tweaks around the edges allowing limetech to concentrate on core functionality.

Is the VM in question set up for UEFI boot? If not, then when creating the VM you want to make sure you select the SeaBios option.

This. It is available under the ‘next’ branch.

It is possible to use .vmdk vdisks directly in KVM but since this is not supported by the Unraid GUI you have to manually put in the full path to such vdisk files to use them.

Unraid does not care about the processor type and is largely hardware agnostic. The easiest thing to do would be to transfer both the USB key and the drives from the old NAS to the new one. It is highly likely that that the system will just come up with the current configuration and drives ready to go.

I know that there have been problems on thetvdb web site this weekend - maybe that is the source of your messages?

The memtest that ships with UnRAID will only work if booting in legacy (non-UEFI) mode. If you want a version that works with UEFI boot mode you have to download it yourself from the memtest web site.

You do not have to worry about scripts being run by User scripts as it handles this correctly for the revised security. If you run your own scripts directly from the flash drive then you have to do something to get them to run correctly.

In theory it should be possible to run the parity check at any time as long as no further errors occur. However I would think it is a high risk as if parity is not good then if any further errors occur you may find that data recovery becomes virtually impossible.

What makes you think those are negative temperatures? Isn’t it just a case of a dash being used as a separator?

If you use the option to Edit the VM then there is a toggle at the top right to switch between form and xml modes.

I have found that there can be times where you need to switch to xml view and remove the usb device. You then switch back to form view and re-add the usb device. Why this can work I do not know but it has several times in the past for me.

You should post your system diagnostics zip file (Tools -> Diagnostics) for informed feedback. My guess is you have corruption on your cache drive but the diagnostics would confirm this.

Sorting out the contents of a lost+found folder can be hard work so one has to decide if it is worth the effort Folders are normally relatively easy as the contents give you an idea of what the folder name should be. Files are much harder once the original name is lost and you have the cryptic ones from the recovery process. If you want to have a go at this then the Linux 'files' command can be useful to quickly determine what type of content is in each file Regarding the message about disk11, the need for the -L flag is quite normal. Despite the ominous sounding warning it virtually never results in data loss and even when it does it only affects the last file(s) being written.

Thanks for that information. It was a classic case of the answer being easy once one has the correct information

You can only run the memtest included with Unraid if booting in legacy mode. If you want a version that can be used when booting in UEFI mode then you need to download it yourself from the memtest web site.

Have you managed to boot to GUI mode on a locally attached monitor? Failure to do so can often mean that the Unraid USB has not been successfully mounted in the second stage of the boot process which is when network drivers and configuration information are read from the USB stick. if you have a console (or ssh) logon you can check on this by issuing the ‘df’ command to see if you have the USB stick mounted at /boot.