Squid

No.

I would think that they mean a zero'ing out / secure erasing of the drive. A low-level format is a misnomer since it hasn't been either necessary or even possible since IDE drives became a thing.

Presumably, the drive you're having problems with is the one you've separately attached SMART for. It IS a 4TB (ST4000VN008)

With the next release of unRaid, then yes. Under 6.8.3 you would use the Unassigned Devices plugin to accomplish the same thing.

I suspect that circus is coming from a container or you've installed it via NerdPack or something. You're going to have to figure out which and then diagnose and/or post in the appropriate application's support thread for more help.

Those USB's suggest poor connections at the motherboard / UPS for the cable. Also try a different port on the server More importantly though, in the middle of all that happening you also have Jun 2 10:55:15 NAS-UNRAID kernel: mce: [Hardware Error]: Machine check events logged You should install mcetools via NerdPack and then at the terminal, post the output of mcelog

https://unraid.net/contact

CPU has to support it. But, with a CPU that "old", even if it did support it (without the modded BIOS), you would more than likely run into a ton of issues with attempting to do any passthrough. Save your hair on this project

According to the logs, it hasn't finished verifying it yet. No idea why that's taking its sweet time. Personally, I never verify as I seriously doubt it would ever come back as a fail.

Assuming that your not assigning the USB to the array, have you installed the Unassigned Devices Plugin & Unassigned Devices Plus and mounted the drive?

That's simply a side effect (who uses Add Container anyways?) They cannot be installed on an unsupported OS

Disk location

Strange. It doesn't populate the autocomplete IF you have a dockerHub (private) conversion present.... Fixed next release

It's working currently for me under 6.8.3 and the release version of CA, along with 6.9-beta and the next rev of CA

Forward the port to Plex and use a reverse proxy for NextCloud. Any docker app only has access (either read/write or read-only) to the files / folders which you allow it access to. While a simple solution to setting up any container on any system is to simply give it carte blance access to every file, IMO there's zero reason to let plex have access to your banking info, so why let it...

The way that any OS works (Windows, MacOS, Linux / unRaid, and the OS within the Krusader App) is that when moving files it first attempts to do a rename to move it, and if that fails then it will do a copy / replace Basically, any move (rename) will succeed within the same mount point. This means that the file stays put on the same disk. IE: Odds on you've passed through a mapping of /UNRAID mapped to /mnt/user Any move within /UNRAID will result in the system simply renaming the file (because all the individual shares are all within the same mountpoint of /UNRAID). Quite normal, to be expected, and will affect every OS ever made If you want to use Krusader to move files from one disk to another then you have to instead mount /UNRAID mapped to /mnt and then move from diskX to diskY BUT, beware that under certain circumstances (moving from diskX to /mnt/user/share or vice versa) you can corrupt the files To be honest, the simplest and fastest way to move between shares is simply via Windows Explorer and cutting / pasting. It is just as fast as using Krusader (the files never actually traverse the network) and will respect the include / exclude settings because you are moving between different mount points (each network share is a separate mount point)

Some unifi routers have an intrusion tester within them (although IIRC the IP addresses in your diagnostics didn't jive with that). Disable that "feature", reboot the server and see if it continues.

Add more cooling. The drives are just sitting there doing nothing other than spinning. During a parity check / rebuild the heads are actually moving, so even more heat will be generated. What kind of temperatures are we talking about?

The parity system is real-time. Any time you write a byte to the array, parity is updated to reflect that. In this case, you had an unmountable drive (disk 2) and instead of running the file system checks against it you checked off the box that said "Format (with the disclaimer next to it), and wiped the contents of the drive. The parity system reflects that disk 2 is now completely empty. How important is the information? Recovery options (EaseUS) do exist to recover the data on the drive by pulling the drive and attaching it to a Windows system. But, without doing that, the system is acting perfectly normal. You formatted a drive, and the data is gone

Depends. You may have installed them within the docker.img itself (only you can answer that) which isn't the greatest idea to begin with. Any updates to the container, and in this circumstance all of your installations will be lost.

Generally yes. The access time is far faster on the cache drive. Easily Redundancy. Personally though, I just backup my appdata from the single cache drive to the array once a week to protect it. Depends upon your own paranoia / how actually important the data is on the array. With a single parity drive, if any drive happens to die then the array is running in a degraded state, and another drive failure before the first is replaced will result in data loss You replace it and the system will build the information again. Also see the above answer. Because of this statement (which you're not alone with) you should not have ANY system on your network directly facing the internet. Whether that's Windows Server / Linux server, if you don't know and understand and have the safeguards in place (and understand exactly how they work) you will odds on make a mistake somewhere along the line and leave yourself wide open to attack. Forward the ports that Plex needs (32400) and you're good. If you do need to remotely access the server, then a VPN (wireguard / openVPN) is your best bet. Reverse Proxies are also ok for accessing apps on the servers.

You made a mistake somewhere along the line. Apps - Previous Apps, Click the download button (down arrow) on piHole and change the template that appears back to where it should be. Alternatively, just search for piHole and install it with the defaults

It checks all the data every time it runs. I run Monthly. You can also install the parity tuning plugin to allow the system to only run for a few hours each night (pauses and then resumes)

Just copy the appdata then