itimpi

If you are parring through a GPU to a VM then it cannot be shared with another running VM. You can only use the sane GPU in two (or more) VMs if they will not be running at the same time. Hardware pass-through is always a bit of a hit-or-miss scenario as it is highly dependent on your exact hardware, BIOS, Linux kernel and KVM versions (most of which are outside Limetech's control) Until you actually try it you can never be sure if it will work for you.

One option might be to start the array without parity2 assigned and tick the parity is valid box. That way you get a protected array as early as possible (assuming parity1 IS valid). You can then stop the array; assign parity2; start the array to build the contents of parity2. You will still want to carry out a parity check afterwards to make sure parity1 WAS valid. The alternative is to make no assumptions about parity and not tick the parity is valid box so that both parity1 and parity2 get rebuilt in parallel when you start the array.

You could always try using a free-standing VNC client rather than the browser based NoVNC version supplied with Unraid. What browsers + OS were you trying to use with NoVNC? Certain combinations seem to have problems.

Reallocated sectors is NEVER a consequence of a bad connection (that will typically result in a CRC error). A reallocated sector means that a write has failed internally to the drive. my suspicion would be that the drive was subject to excessive shock prior to it reaching you and that has caused internal damage to the drive. The drive also failed the extended SMART test which is invariably a sign that is not in good shape.

I would get that drive out of your Unraid system ASAP as it looks to be in bad shape. Although small non-Zero values for reallocated sectors in itself is not worrying, it is VERY worrying if not stable. Any value other than 0 for pending sectors means the drive cannot be read reliably, so if another drive fails the rebuild is likely to have as many corrupt sectors as there are pending sectors on other drives (a good rebuild requires ALL non-failed drives to read without error). finally any drive that cannot pass the extended SMART test should be considered failed.

I believe that all browsers on iOS use Apple’s WebKit as the underlying rendering engine so that may well mean all iOS and iPadOS browsers will have the same issue.

You might want to look into using the Parity Check Tuning plugin to help with only running a parity check outside prime time.

The WireGuard VPN server built into Unraid 6.8 provides this sort of capability.

The normal way to stop this sort of event potentially causing a problem is to set the router to reserve the IP address that Unraid is to use. That way you can be sure that no other device on the LAN can be given that address

It should be the other way around - you should not be letting any ports through to Unraid that you have not explicitly decided that it is safe to do so.

The end-user cannot realistacally install drivers into Unraid. It has to be done at the Limetech end.

How are they not working? What formats are they? Are you aware that Unassigned Devices is now split into 2 plugins. The basic one provides the commonest features while if you install the Plus variant as well you get some additional options added.

It is normal for loop0 and loop1 to be full as they are files loaded by Unraid as part of the boot process. Loop2 (the docker image) being full is very unusual. It may well mean that a docker is mis-configured and is writing internally to the docker image. Normally docker containers are configured so that all variable data is mapped externally.

The Previous Apps option DOES have the settings last used if you re-install a docker container from there (assuming you mean all the mapping of shares, ports etc).

If you enable Netbios under Settings->SMB then according to the built-in help this also enables SMB1.

Any time you make a change then the WireGuard service will be stopped and then restarted. I myself have noticed that it "sometimes" seems to leave the service inactive, but I have not found a pattern to reliably reproduce this for further investigation.

As was said if you want files on the cache to be protected then you need at least 2 drives as a cache pool in the cache configured as RAID1 (the default for 2 drives) to give redundancy. An alternative strategy is to leave them unprotected but use a plugin like CA backup to make periodic backups to the array.

Not that I know off! The drivers and service are built-in. The GUI component for configuring it is currently a plugin so that it can be easily tweaked. I was using it very successfully over the Christmas period to access my Unraid server (and other devices on my home Lan) remotely.

No idea then am afraid. Everyone else seems to get past this step. just had a thought - what size vdisk did you create? If the size is too small I think you can get strange error messages.

I have no idea why you are not able to see the virtio drivers, but you could try changing the vdisk type to SATA from virtio in the VM definition which should allow you to get the VM running without the need to install the virtio drivers during install.

Are you sure the full path inside the container is really /images (i.e. it as at the top level of the container file system) ? The code you posted has no leading / on the name which means it is a path relative to the current folder. When you console into the container you might want to check with the pwd command what is the current folder.

I would suggest you post the system diagnostics zip file (obtained via Tools->Diagnostics) which might give some clues.

If you install the Parity Check Tuning plugin then it adds this information to checks it is controlling/monitoring (as well as the other features it provides).

If the drive has been manually formatted then it is difficult to see how this error can occur unless the USB stick is in read-only mode for some reason. Can you copy ANY files at all to the USB stick(s)?

The FCP (Fix Common Problems) plugin can do quite a bit of this. I can see this best to be kept as a plugin rather than built-in so it can more easily be tweaked/enhanced.