Squid

6.10-rc3+ will not update the templates for you (when it's released SOON(tm) On previous versions of the OS, edit the applicable xml in /config/plugins/dockerMan/templates-user on the flash drive and change the line that says <TemplateURL>someURL</TemplateURL> to instead be <TemplateURL>false</TemplateURL>

If you have a supported system (https://github.com/liquidctl/liquidctl#supported-devices) you might want to look at

Yeah, somethings holding it up. Do you have an attached monitor / keyboard to the system? Reboot should take around 5 minutes max depending upon what you've got installed

Its a memory thing that's the cause (presumably) (or mc doesn't support gz that large). Not currently possible to separate things out, but it's on the drawing board. I'd suggest opening it via WinRar on a Windows Box

Why wouldn't you?

Reseat all the cabling to the drives (power and sata) and be careful not to disturb anything you've already reseated. It looks like just a very poor connection

Actually, the mce happened during processor initialization and happens on certain hardware combinations and is perfectly safe to ignore.

You're running ECC memory. The memtest that comes with the OS will not display any errors because they're being corrected. Does your system event log in the BIOS tell you anything more. Set up a new bootable stick via https://www.memtest86.com/ and run it (it's supposed to find ECC errors)

You should post your diagnostics

It because in reality there's no such thing as isolating a core from the OS. What you're doing in actual fact is stopping the context switcher from utilizing certain cores. It's the context switcher that dictates which cores which process can run on. At the end of the day this results in a process being able to run on any core that's not isolated and on the lowest numbered isolated core. For something like Plex I wouldn't pin it to any core. Pinning makes things more confusing (hence these posts), and by not pinning it you're giving it access to all the non isolated cores.

Docker containers are part of the OS. Therefore if you isolate the cores from the OS, the docker containers will NOT run how you think they should if you pin them onto cores the OS does not have access to. In you case, the OS only has access to cores 0 & 8. You've told Plex to run on cores 1 through 15 This means that Plex is only going to run on core 1 (the lowest isolated core) and core 8 (the non isolated core). It will not run on anything else Simply put, only isolate cores if you're using those cores specifically for VMs. The over simplified way to think of it is that if you isolate the core from the CPU is that nothing other than a VM can use it. period.

VMs should for best performance be on isolated cores (but entirely up to you). If @SpaceInvaderOne suggested to put a docker container onto isolated cores, then that's a mistake. In particular the screen shot at 8:46 is incorrect The OS will run on cores 0 and 4, and plex will run on 0, 4, 1. It will never run on 5 because both 4 and 5 are isolated from the OS. And with multiple isolated cores pinned to a container it will only ever run on the lowest isolated core. FCP is correct.

Yes, that isolates every core from Unraid except for the first one. But you've got Plex running on the isolated cores. You have to bear in mind that docker is a subsystem of the OS, so isolating a core from the OS effectively isolates it from docker. CPU isolation is generally only used for VMs

Changed Status to Closed Changed Priority to Annoyance

You're attempting to pin Plex to cores which you've isolated FROM the OS in Settings - CPU Pinning. So if you've isolated cores 3-6 from the OS and decide to pin Plex to those cores, Plex will only ever operate on core 3. If you don't isolate those cores from the OS, then Plex will get pinned properly onto cores 3-6 There is no limitation per se on what the number of cores Plex can use. It's simply how you've set up your isolation which will give results that you aren't expecting, hence the FCP warning A very safe assumption is that no app can handle being pinned to multiple isolated cores without only executing on the first core. The "exactly know what you're doing" implies that you're going to be manually setting the affinity for certain processes by hand etc, which no one other than a developer themselves is ever going to do.

All the UEFI check box does is rename the folder on the flash drive from EFI~ to be EFI

Delete the file

That looks to be actually corrupted. Not to mention that /boot/config/AAAremaneddotsmart-one.cfg isn't really the proper filename for that file in the first place. Delete it

Huh. Learn something new everyday

A parity check running would be a dead giveaway

Post your diagnostics

rename the file to be docker.img instead and try again. I seem to recall an issue with that.

First I've ever heard of it. Post your diagnostics

Upgrade to 6.10-rc2+ Or alternatively disable in Settings - Notification Settings dynamix checking for plugin updates. With it enabled,l 6.9 automatically checks prior to loading the page Operation of how this all works is different in 6.10+