itimpi

Does Hyper-V support hardware pass-through of USB flash drives? Unraid would require this as it both boots from a flash drive and uses the flash drive GUID to validate the Unraid licence. If this is supported then it might be a good way to get an Unraid VM running to try out its capabilities before committing too many resources to it.

😁

DiskSpeed (the one that seemed to have crashed for you)

Missed the fact that the diagnostic had no SMART info for disk5 suggesting it had dropped offline.

Not sure what can cause that☹️ just checked and that container is working fine on my 6.8.0-rc1 system. I wonder if it is related to your current issue in any way - the fact it mentions a timeout does feel it could be? Might be worth making a post in the support thread for that container to see if the maintainer has any ideas. BTW: I looked at the diagnostics and did not spot any obvious problem.

Does that mean it is likely to remain an outstanding issue? Just like to know for future reference i am not too worried as I can easily work around it by using a different browser or puTTY to get to the Unraid command line. If it remains an issue you might want to add it to any future release notes.

You should post your system diagnostics zip file (obtained via Tools->Diagnostics) to see if we can spot any reason for the slowdown. If the diagnostics show no obvious reason you can also consider using the DiskSpeed docker container to check if any of your drives have ‘slow’ spots that might cause such symptoms.

What subnet is your phone on when this happens, and what are the subnets for for the Unraid server and the dockers? I think I saw a report of similar symptoms when the remote subnet and the subnet used by Unraid were the same as this resulted in a routing issue. If this is the case then it is going to need looking into for the most robust solution as the user rarely has any control over the remote subnet. If course it could be my memory is faulty and your issue is something completely different

It I worth checking your drives to see if you have an ‘appdata’ folder with same spelling but different capitalisation (e.g. Appdata). Linux is case sensitive for file names while Samba (shares) is not. In the case of there being folders with the same spelling but different capitalisation at the Linux level then only one of them will be visible at the samba (share) level.

Because that ‘yes’ refers to where NEW files are put, not to where existing files are located. Reading the GUI Help text for this option might make that clearer.

For this purpose ignore the /mnt/user* variants as they are just different views of the files on the physical disks being seen via the User Share layer in Unraid. I think the 'cache' variant of the file takes precedence at the User Share level. If /mnt/cache is not working then try the /mnt/disk1 path in case that is the one that contains your docker images.

You can also get that type of indication on the Dashboard if the CRC error count has increased. CRC errors do not show up when you run a SMART test as they indicate connection issues rather than disk failures. In such a case clicking the Orange icon and selecting the Acknowledge option will make the icon turn green until another error occurs. The CRC count never resets to 0 so Unraid only notifies you again if it increases.

If you think about it the mount must happen before the 'go' file executes as otherwise the 'go' file would not be found in the first place In fact the mount must happen very early in the boot sequence as the flash needs to be mounted to enable Unraid to read all its configuration information.

Why bother? You can always use the 'go' file to copy the scripts to a location from which they CAN be executed.