john_smith

does healarr support self-hosted ntfy notifications ?

I did some testing before shutdown and in my case it looks like its the /mnt/user directory for me show up as a loop in the syslog: root: rmdir: failed to remove '/mnt/user': Device or resource busy after unmounting it directly (umount -l /mnt/user), the loop blocking stop of the array ended and it stopped

How did you determine it was that specific folder? I'm curious to find which is the issue for mine, as I'm also using zfs

I've discovered i have the same issue over the past few updates, since i never reboot otherwise and haven't had any other issues. The last 2 updates I've stopped all VMs and docker containers fully, spun down disks, and then rebooted and both times it still get stuck on the same spot and then kills it and calls it an "unclean shutdown" and finds no array errors (since nothing was running or being written to the array). Reading this thread, I'm still not clear what the issue is though. @sasbro97 did you ever figure it out or did it resolve itself when the cloud folder in your case was moved/deleted?

back with another comment - i don't think a diag report is needed here as this is more aesthetic on the new feature the new privacy toggle is a great idea and i love it - that being said, it's a bit useless in its current iteration for two reasons: visual obfuscation is easy to deconstruct, even just visually if you know what icons look like, but moreso with an actual tool IPs are not obfuscated. Only ports should be visible ideally, or perhaps add the option to select what we want blocked: container IPs local IPs ports

@XCrownedClownsX FYI just tested this again, on both firefox and edge incognito, same issue persists. here's the support bundle.FolderView Plus Support Bundle-2026-04-06T19-19-17+00-00.json

FYI, at least part of the issue I mentioned recently is still present as of the current version. When the docker tab is loaded with a folder that has at least 1 container with an update minimized, and that folder is expanded, all of the container in the folder show the "apply update" update button. Loaded minimized: Expanded: Refreshing the already expanded folder, you can see this does not happen if the page is loaded with the folder already expanded:

there seems to be a bug with how the "apply update" button is showing up, both visually and in function (pressing it), on the current version. When a folder is expanded for a folder that has at least 1 update pending, it shows ALL containers have an update pending. Only 3 had updates pending. I only wanted to update one, and i clicked apply update for the specific one - however, the button is actually just mirroring the folder-wide apply update, and went ahead and started updating ALL CONTAINERS in the folder (all 3 in this case). After the updates were applied, the folder-wide "apply update" still did not go away either This is not what i would expect, both visually and functionally

does this support autoheal?

@SpaceInvaderOne i looked at a similar plug-in to yours (a version more complex than I'd prefer tbh), and noticed during attempted setup that of my disks happens to not use the same encryption key for some reason as the others. I assume this is why your plugin didn't work for me out of the box, but I'm not sure why i have no issues with manual decryption. Have you run into this issue? How would I fix this to be able to use your plugin?

anyone have one of the other repos installed as well to facilitate import of data? or otherwise some other way of syncing? wondering how to do this part now that i have the container installed

If they're harmless, why are the messages coming up?

discovered this just now, can someone elaborate the difference between using just this and appdata backup plugin? i already have duplicati setup, but wondering what this adds

Recently replaced my Coral A+E m.2 TPU with a Hailo-8L B+M m.2 TPU, which has made Frigate stable, with the TPU itself never going offline in the weeks since. However, today I noticed the following error keeps popping up associated with the kernel, such that when it does it happens numerous times that day, but then it won't happen for at least a week or so. Is this a concern? Log excerpt here

I had reseated it a few times, and never saw the temps go anywhere near even throttling. I don't know whether it was a power issue under load, something buggy with the kernel/driver/apex, or a PCIe conflict/instability, but tbh it sounded like more than one of these, and I was a bit tired of troubleshooting this. Given all i've read about the issues with coral A+E key TPUs causing issues on desktop motherboards' m.2 slot meant for wifi chips (here and elsewhere) I ended up replacing my coral M.2 A+E key TPU with a Hailo-8L M.2 B+M key TPU. It's only been a few weeks but haven't had any issues since.