Squid

Review this https://wiki.unraid.net/Upgrading_to_UnRAID_v6

Some combinations of hardware (cpu / motherboard) will issue a MCE when initializing the CPUs at boot up time (Anecdotally, I've also determined that this can also be caused by ignoring your wife telling you to pick up your socks). This is normal, and is what you're seeing, and safe to ignore (on the MCE being issued). But, the ryzen tweaks / fixes linked above still need to be handled.

Make a note of what your parity disk(s) are. Also, if you are only using a single parity disk and its assigned to the first parity slot, then when starting the array, you can also check off Parity is Already Valid. If you are using 2 parity disks (or only one and it's assigned to the 2nd parity slot), then you will have to rebuild the parity information (ie: Parity is NOT valid)

What's the command in the go file?

Did you install the plugin with the proxy running?

I know. It's hard to remember everything you do. But, as an aside, shortly after that patch came out, the auto-update plugin incorporated it (in a safe manner), and the patch was no longer necessary. It's more a kind of "kick Squid in the ass" because it never occurred to me to even look at that file in the go until after everything else was done.

If the Update option is present in the context menu when in the Docker tab, then this is a known issue and fixed for next release of the OS

No idea why you have this line in your go file (/config/go on the flash drive), overwriting the stock file, but it's going to be the cause yes | cp -f /boot/config/root/DockerClient.php /usr/local/emhttp/plugins/dynamix.docker.manager/include/DockerClient.php

Rather suitable now

The only thing that even begins to make sense to me is a corrupted flash. Download the zip file for 6.8.3 from limetech's website, and overwrite all of the bz* files on the flash drive with the ones in the zip. Then reboot.

Try restarting in Safe Mode from the boot menu to see if that makes a difference

Have you ever formatted the cache drive? It's showing that it's not formatted

You could always run the procedure manually Format a stick as FAT32 Make sure that its named UNRAID Download the zip file from Limetech's website Extract the contents of the zip onto the flash drive Right click on make_executable and run it as administrator done

I would reboot and see if anything changes. If not, then post your diagnostics. ie: There's no reason why that error would happen...

More accurately, it would be that the fstrim command itself thinks that the drive(s) are SSDs. Since you can't do anything about the command itself, what you'd have to do is forget about the plugin, and run the appropriate commands via the user scripts plugin on an appropriate schedule. IE: fstrim /mnt/cache -v Would only trim your cache drive

Feel free. There's only 1 ground rule. Criticisms aren't allowed (CA's been a 5 year project, and while its super stable, if I was going to do things all over again, it would not be programmed the way that it is (especially the code from the first year or so)

FYI, all of Sparklyballs' containers were removed completely from CA because of zero support, they were all deprecated, and a large chunk of them were no longer operational.

Yeah that was intentional. Those first 5 do more processing at startup that isn't conducive to any of the other sections, and the other sections pinned / installed / previous may break without the extra processing needed. After an install however, you will always wind up wherever you were.... A fix isn't impossible, but unlikely I'll do it anytime soon.

^^ This. But, I wouldn't worry about it with a 20G docker.img file

Mar 23 05:06:36 BisecTechUnraid root: Data CACHE Level-1 Data-Read Error This implies that the L1 cache (ie: CPU) is starting to go bad. But, it is also entirely possible that you got hit by a stray cosmic ray. Reboot and monitor to see what happens. (It only happened once in the diagnostics - 3/21 02:09:04)

They would direct the appdata share for those affected apps to be /mnt/diskX/appdata/... Without really diving into all of the release notes and comments scattered throughout, under global share settings is a Use Hard Links option. This IIRC was added in (according to the help text) to improve compatibility with NFS, but I believe that the option also affects how the inode numbers are handled on hardlinked files. As itimpi alluded to, this has been an ongoing issue for some containers and some users. You will see the occasional comment scattered throughout the threads with the advice being given to a user to switch to /mnt/cache/appdata. The trouble however is that for other users leaving things at /mnt/user works no problems. And, despite some occasional posts from myself to the docker authors to the effect of "OK, fine what is the actual problem", no one has thus far been able to point a finger at "this is why its failing" If anyone could ever outright point out why /mnt/user fails for some but works for others, then I'm sure Limetech would fix the issue. The advice I'm offering you, since you've already diagnosed that the app wants /mnt/cache, is to see whether or not it works as /mnt/diskX If it does, then basically you're good to go. If it doesn't, then you've got to get together with Limetech and figure out what's going on.

You can install the dynamix schedules plugin to adjust the time of daily weekly monthly scheduled tasks

Yeah, it's deprecated in favour of the lsio boinc app. It does still work however, just its an older version of boinc within it

Personally, given the amount of times this comes up regarding Plex, I'd seek assistance in their support thread or on reddit. Nothing against this forum, but at first glance, this is going to be a Plex issue itself, and the other support venues may yield faster results. All this assumes of course that within Plex's UI that you can still access the library (or it's still adding in new media files from the library)

Misread your post a bit. Plex's config files for its listening port would be stored somewhere within the appdata share for plex. I don't use Plex, so can't offer any further advice than that though. You would probably be able to ask and get a response on Plex's own support forum for that kind of detail since they do offer their own docker container.