Squid

Not a bug, but something to be aware of if you're really playing around with 6.8 and aren't understanding what you're doing. If you happen to change the landing page for unRaid to a custom tab that is set to URL (open same tab), then for all intents and purposes you will have effectively locked yourself out of the server's GUI and will have to manually edit the appropriate config file on the flash drive to restore the landing page back to something normal.

May not be exactly what you want, but you can change the "start page" the system takes you to via Settings - Management Access (You can select any tab for the default, including any custom ones you may have)

The safe thing is to redo from scratch, and simply add on any custom path mappings etc you've made to the template. The replacement will probably use the same /config folder so maybe there won't be any major issues. YMMV The version from xamindar may get removed from the apps tab in the near future.

You can change the threshhold by Main, Cache Devices and clicking on Cache.

Or bad memory (run a memtest), as the file being referenced in the OP's screen shot is in RAM

Oh, and you can also always go to the appropriate support thread for any given application by clicking on the app's icon and then hitting "Support"

Since it *appears* that you are using syncthing from xamindar, I'd suggest possible changing to syncthing from linuxserver (also available in apps). The latter supports PUID and PGID settings which in theory should write files with more appropriate permissions.

Docker Tab, hit Container Size at the bottom. Will tell you pretty much everything you need to know about where it's coming from and then you can make appropriate adjustments with your logging, path mapping etc

Plugins dependent upon other plugins. (Mainly notes for myself so I know what CA is doing) Plugins at boot time are installed in order of the ASCII sort (not natural) of the filename including the .plg extension CA when doing multiple previous app installations installs them in order of an ASCII sort excluding the .plg extension. This handles the special circumstance where UD+ looks for UD to be queued for install at boot prior to installing but it actually winds up installing first at boot time. If you make a plugin that is dependent upon another plugin being installed and it errors out if the other plugin is not installed, you must ensure that you can handle both very slightly differing sort orders.

Actually, yes. Kinda threw me for a loop on CA because I was expecting javascript to sort the array as so that UD+ would come after UD. But, the reverse is actually true (had to have CA do a reverse sort) And then looked at my system log to see which order things are actually installed in from a boot and it's not what you expect, and your change won't work. Jan 11 07:50:21 ServerA root: ----------------------------------------------------------- Jan 11 07:50:21 ServerA root: unassigned.devices-plus has been installed. Jan 11 07:50:21 ServerA root: This plugin requires Dynamix webGui to operate Jan 11 07:50:21 ServerA root: Copyright 2015, gfjardim Jan 11 07:50:21 ServerA root: Copyright 2016-2019, dlandon Jan 11 07:50:21 ServerA root: Version: 2019.12.29 Jan 11 07:50:21 ServerA root: ----------------------------------------------------------- Jan 11 07:50:21 ServerA root: Jan 11 07:50:21 ServerA root: plugin: installed Jan 11 07:50:21 ServerA root: plugin: installing: /boot/config/plugins/unassigned.devices.plg Jan 11 07:50:21 ServerA root: plugin: running: anonymous Jan 11 07:50:21 ServerA root: plugin: skipping: /boot/config/plugins/unassigned.devices/packages/unassigned.devices-2020.01.10a.tgz already exists Jan 11 07:50:21 ServerA root: plugin: creating: /tmp/start_unassigned_dev UD+ installs at boot before UD. EDIT. nm you're looking at /boot and not /var/log/plugins to see if UD is there

Wonderful Update to CA will be required so that on an multi-install from previous apps it sorts the plugin list before installing in case you checked off UD+ before checking off UD

While nice, way too complicated and won't get implemented. The plugin does however offer the option to not stop any container, (or not stop selected containers)

There's also a network-routes file or similiarly named within the config folder you can also delete. But, if you didn't reboot then nothing changes...

The exact same URL as the what appears in the address bar if you hit the webUI from the docker page for Krusader

That is exactly how you do it. Whether or not the file gets recreated until a change is made to the settings doesn't matter as unRaid will utilize its defaults in the interim.

Just in case people do miss it (and they probably will), the policies regarding CA and the applications contained are now publicly posted. It's more just iterating what myself and @dockerPolice / @pluginCop already do on a daily basis. (And these policies are impartial. If anything, @pluginCop is hardest on me - the guy's just got it in for me for some reason )

And you will also need to run the file system check on disk 2. That corruption is usually caused by unclean shutdowns if the system is in the middle of a write Jan 9 22:40:09 Tower root: Fix Common Problems: Error: /boot/config/plugins/dockerMan/templates-user/my-tautulli.xml corrupted The flash drive is also corrupted.

Reset the BIOS?

First thing would be to reconfirm the boot order in the bios

I would just keep using SMB and windows for it. Sweet and simple

Just tested it, and dockerHub templates (if you did the search and add via CA, or by manually filling out a template completely) will show up in previous apps. The only thing is that IF you had one, named it AppX and then installed something else and also named it AppX, then the first one would be gone. Limitation of the underlying system. Anything created via a command line though would never be saved as there was no template to save in the first place

Works for me Only thing that springs to mind is that you're accessing unRaid via https, which means that no iframe can be http. But that doesn't make a ton of sense as you wouldn't see anything at all. I'd try a different browser / clear cache / etc

Dead board? But, Marvel based controllers are also no longer recommended due to the continual problems as time goes on. Best off switching to a LSI based one...

What @trurl said, but this is the key line in everything you just listed (And unlikely anything is actually lost)

Does Krusader still work normally?