Squid

The path has to end in a slash on this version (And has to be a non-existing directory for the initial build)

Jul 21 05:33:13 Blackbox kernel: mce: [Hardware Error]: Machine check events logged Jul 21 05:33:13 Blackbox kernel: EDAC sbridge MC1: HANDLING MCE MEMORY ERROR Jul 21 05:33:13 Blackbox kernel: EDAC sbridge MC1: CPU 1: Machine Check Event: 0 Bank 5: 8c00004000010093 Jul 21 05:33:13 Blackbox kernel: EDAC sbridge MC1: TSC 11995fad7bbb2 Jul 21 05:33:13 Blackbox kernel: EDAC sbridge MC1: ADDR e17739bc0 Jul 21 05:33:13 Blackbox kernel: EDAC sbridge MC1: MISC 2042765686 Jul 21 05:33:13 Blackbox kernel: EDAC sbridge MC1: PROCESSOR 0:206d7 TIME 1595302393 SOCKET 1 APIC 20 Jul 21 05:33:13 Blackbox kernel: EDAC MC1: 1 CE memory read error on CPU_SrcID#1_Ha#0_Chan#3_DIMM#0 (channel:3 slot:0 page:0xe17739 offset:0xbc0 grain:32 syndrome:0x0 - area:DRAM err_code:0001:0093 socket:1 ha:0 channel_mask:8 rank:1) Memory going bad

One of the VM's caused the system to kill off processes, with unpredictable results. (They are currently the biggest users of the CPU and memory). The main problem is that the load average is skyhigh which is killing all responsiveness to eveything.

Doesn't particularly matter. If you're running dual parity drives on that system, then your speed may be slower when running a parity check / rebuild. With only a single parity drive, it doesn't really matter how many drives are present. (All subject of course to bandwidth limitations on the controllers, PCIe slots)

The only way that has been found thus far to get an unitialized csrf_token is if (probably due to a misconfigured docker app) the RAM (rootfs) completely fills up. The diagnostics are after a reboot, so not much of any use there. Watch the dashboard, and when the log / used memory goes up significantly, post new diagnostics.

As a file server, with light docker container usage, then yes that's sufficient.

Pitch it. Regardless if it works or not, or is sufficient in terms of wattage, the "generic" P/S (especially the ones that used to come as a freebee with a case) are complete garbage. The P/S is the single most important component in any system, as everything else depends upon it. Yes. The parity drive though has to be as large or larger than the largest data drive. Yes. But, for SATA expansion cards, it is best to avoid any card which is Marvel based. LSI / AsMedia would be the go to's Yes

Aug 5 01:09:44 Tower kernel: mce: [Hardware Error]: Machine check events logged Aug 5 01:09:44 Tower kernel: EDAC sbridge MC0: HANDLING MCE MEMORY ERROR Aug 5 01:09:44 Tower kernel: EDAC sbridge MC0: CPU 0: Machine Check Event: 0 Bank 7: 8c00004000010090 Aug 5 01:09:44 Tower kernel: EDAC sbridge MC0: TSC 40a13e7678cb1 Aug 5 01:09:44 Tower kernel: EDAC sbridge MC0: ADDR f380c8640 Aug 5 01:09:44 Tower kernel: EDAC sbridge MC0: MISC 1402e0086 Aug 5 01:09:44 Tower kernel: EDAC sbridge MC0: PROCESSOR 0:306e4 TIME 1596600584 SOCKET 0 APIC 0 Aug 5 01:09:44 Tower kernel: EDAC MC0: 1 CE memory read error on CPU_SrcID#0_Ha#0_Chan#1_DIMM#0 (channel:1 slot:0 page:0xf380c8 offset:0x640 grain:32 syndrome:0x0 - area:DRAM err_code:0001:0090 socket:0 ha:0 channel_mask:2 rank:1) Bad memory. The system event log in the BIOS would hopefully provide a little more insight to which DIMM is starting to go, besides Channel 1, Dimm 0

False positives related to FCP and cache-pools will get fixed. Sorry... Didn't even think about FCP's test on that one when I started with beta25

Try 2 tests. Leave it on the dashboard (which it appears you've been doing), and then leave it on something like Settings

Are you sure you're moving to & from the same mount point (ie: /media to /media), or is it different mount points (/UNASSIGNED to /media)?

Does it always work fine with another browser?

And post your diagnostics, since on the surface everything seems fine...

And that's what the -P on the pkill is supposed to do. Kill off children from the main PID

But was the script itself still running? It's a royal pain to kill off sub-processes from the main script, especially since the only thing I've got to go on is the name of the main script. The system does attempt to kill off children (pkill -TREM -P). If you can come up with a better solution, then please let me know.

Dynamix streams. They're similar at a high level, but the streams plugin is more limited in that it only shows what is actually streaming over the network whereas this plugin shows everything.

Most people (I would hope) use the appdata backup plugin to make a copy say weekly for this circumstance. I would suspect that your cp command is going to wind up massively exploding the size of Plex's appdata (holiday long weekend so not going to try) due to plex making extensive use of symlinks but it won't in and by itself cause any problems

That test is done by scanning the syslog and seeing if the cpu has throttled itself down at some point. It is not a "live" test. If you have that warning the cpu has throttled down because of temperature since the last reboot. Up to you if you ignore it or not.

And what's the gateway and dns set to?

Nothing to worry about and ignore it. For some reasons, certain cpus and motherboards issue an mce when initializing the cpu cores.

How would you rectify that? There is the little info icon for every plugin on the plugins page, and for those plugins which support it (the dynamix ones do not unfortunately), if the update plugin banner comes up then there's also the info icon present there.

macvlan traces I *believe* have been minimized with 6.9 beta25+ Also, since you mentioned "tainted", that word means something not quite as ominous as it sounds. Basically, it means that you're not running a stock linux kernel (ie: you're running unRaid).

You're running a backup of the appdata at 3am. Part of the standard process in the backup is (unless you tell it otherwise via advanced settings) to stop all of the containers. They will restart when the backup (and optional verify) is completed.

You've probably got a display port on the video card in addition to the HDMI. Pick up a display port -> HDMI adapter for the 2nd display