trurl

The license key file is for the specific flash drive you registered it on. It is associated with the GUID of that flash and won't work on any other. You can do a license transfer from one flash drive to another. This results in a new license key file for the new flash, and blacklisting the old flash GUID.

Unless you reinstall using the Previous Apps feature on the Apps page it will use a fresh template.

On mobile now so can't look at Diagnostics. Have you done memtest?

The paths an application writes to is set within the application.

Filling docker image is typically caused by an application writing to a path that isn't mapped.

In addition to parity errors a lot of this in syslog: Aug 7 12:51:35 Quetzalcoatl kernel: CPU1: Core temperature above threshold, cpu clock throttled (total events = 27714) Aug 7 12:51:35 Quetzalcoatl kernel: CPU5: Core temperature above threshold, cpu clock throttled (total events = 27714) Are you overclocking? Why have you given 200G to docker.img? 20G should be more than enough and if you fill that you have some application misconfigured.

Google seems to suggest it is DVD. SMART for disk2 is missing in those diagnostics and syslog suggests connection issue. Emulated disk is mounted though so would expect rebuild to be OK if you fix your connection issue. Looks like you have Marvell controller. Lots of reports of problems with those on Linux. But some people have success with those if they disable IOMMU in BIOS. Do you have any VMs that require passthru? Why have you given 100G to docker.img? 20G should be more than enough and if you fill that you have some application misconfigured.

https://wiki.unraid.net/The_parity_swap_procedure Note the array will be offline during the "parity copy".

Parity is simply an extra bit that allows a missing bit to be calculated from all the other bits. That is basically all it is wherever it is used in computers and communications. All parity check can tell is that the calculated parity from all the data disks doesn't match parity. No way to know from just that extra bit which disk might be out of sync. If you have no reason to suspect any particular disk, you have no choice but to correct parity. SMART for all disks looks OK, and nothing about any problems communicating with any disks in that syslog. However, there is this: Jul 26 12:28:20 Tower emhttpd: unclean shutdown detected That number seems reasonable for unclean shutdowns. Why did you have an unclean shutdown? Unclean shutdown results in non-correcting parity check. You have to run a correcting parity check to correct the parity errors it found. Then you should run another non-correcting parity check to make sure you don't still have some problem. Exactly zero parity errors is the only acceptable result, and if you don't get there you aren't finished.

Lots of this in syslog Aug 6 04:38:27 Unraid kernel: CPU3: Package temperature above threshold, cpu clock throttled (total events = 9521810) Aug 6 04:38:27 Unraid kernel: CPU7: Package temperature above threshold, cpu clock throttled (total events = 9521810) Aug 6 04:38:27 Unraid kernel: CPU0: Package temperature above threshold, cpu clock throttled (total events = 9521810) Aug 6 04:38:27 Unraid kernel: CPU4: Package temperature above threshold, cpu clock throttled (total events = 9521810) Aug 6 04:38:27 Unraid kernel: CPU5: Package temperature above threshold, cpu clock throttled (total events = 9521810) Aug 6 04:38:27 Unraid kernel: CPU1: Package temperature above threshold, cpu clock throttled (total events = 9521810) Aug 6 04:38:27 Unraid kernel: CPU2: Package temperature above threshold, cpu clock throttled (total events = 9521810) Aug 6 04:38:27 Unraid kernel: CPU6: Package temperature above threshold, cpu clock throttled (total events = 9521810) don't usually see that so much. Are you overclocking?

I let everything on my network use DHCP, and if I want specific IP addresses for certain devices, I reserve that IP by MAC in my router. That way it is all managed in one place.

As for whether or not other things in your configuration could be improved, post your diagnostics and we can take a look. Go to Tools - Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread.

Assuming user shares have default Allocation Method of Highwater, what you have is normal. You can adjust the warning level for each disk as seen in that 2nd screenshot you posted. Disk1 has 4TB free, so disk1 is chosen first, and Highwater is set at half the remaining capacity of the disk, 2TB free. So it uses disk1 until it has 2TB free. Then it re-evaluates Highwater. Now disk1 has 2TB free, and all other disks also have 2TB free, so disk1 is chosen first, and Highwater is set at half the remaining, 1TB free. It will continue to use disk1 until it has 1TB free, then go on to disk2 until it has 1TB free, and so on. Other allocation methods are available, but Highwater is default for good reasons. It is a compromise between using all disks (eventually) without constantly switching disks, as would happen with Most Free Allocation.

Go to Tools - Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread.

Not clear if it would help but your RAM is a bit tight. Can you add any? You are probably going to need more if you intend to run many dockers anyway.

This thread is mainly a support thread for this container as used on the Unraid platform. This is the Unraid forum.

Instead of linking to unknown external sites, please attach directly to your post. Also, when posting your docker setup, instead of screenshot of that page, it is better and more complete to post the docker run command as explained at this very first link in the docker FAQ:

The log button on the menu bar displayed on every page. If it isn't working for you maybe you have something blocking it. Whitelist your server and/or try another browser.

Have you tried another port on your router / switch?

https://forums.unraid.net/topic/86066-boot-fails-post-68-upgrade-wkernel-panic-new-usb-stick-not-booting/?do=findComment&comment=797764

Are you referring to the Recycle Bin plugin? If so, that suggests it is being deleted over the network from another computer.

Just to close the loop on this, you need to tell the application to use /data, which is the container path. The application can't know anything about the host path, /mnt/user/Downloads. All dockers work this way. Understanding volume mapping is 90% of knowing how to setup any docker.

Does it happen when not running the NVidia build?

A lot of discussion about adding a drive to a new slot, but possibly more common would be replacing a disk with a larger disk. The larger disk is rebuilt from the parity calculation, which includes that space on parity that previously was beyond the size of any of the disks.