The_Holocron

Hey @jcofer555 , thanks for the update on the "depreciated." On a separate topic, if you want assistance with maintaining this, I'm open to helping.

Curious, you mention it is basically in "Maintenance Mode" in the description: So why set it to depreciated? Because now if you keep it installed you get the constant nag to uninstall it.

I fixed it by changing the value in the template. It is what was noted above.

Permissions, right. Always permissions. I will need to check that. Log: ✓ Database initialized GyroidVault running on http://0.0.0.0:3000 Starting background library scanner (interval: 24 hours) [Scanner] Start library scan: /app/data/library [Scanner] Scan voltooid. Modellen: 0, Bestanden: 0, Overgeslagen: 0 [Scanner] Start library scan: /app/data/library [Scanner] Scan voltooid. Modellen: 0, Bestanden: 0, Overgeslagen: 0 [Scanner] Start library scan: /app/data/library [Scanner] Scan voltooid. Modellen: 0, Bestanden: 0, Overgeslagen: 0 ✓ Database synchroon weggeschreven (shutdown). ✓ Database initialized GyroidVault running on http://0.0.0.0:3000 Starting background library scanner (interval: 24 hours) [Scanner] Start library scan: /app/data/library [Scanner] Scan voltooid. Modellen: 0, Bestanden: 0, Overgeslagen: 0 I figured it out. The template was still pointing to the /library and not /app/data/library internally as noted in the post above:

Library scan not seeming to work. I have it mapped to an NFS remote share that has all my 3D files, but scan library shows nothing.

Things are looking pretty good. Thanks for all of this. One thought I just had is to have a "Manual" selection in the frequency selection. This would allow you to set up a backup plan that would only be manually triggered to run. You can basically do this by setting up a scheduled backup, disabling it, and only ever clicking on RUN. Maybe there's an alternative way to set this up.

I'm not seeing this in the UI when a backup is running. Also, 12 hours to backup my VMs. Wow...that's a long time.

Upgraded to 7.2.4. Unable to save any scheduled backup or update existing. I uninstalled and reinstalled with same issue after. Backup Now does work.

BTW, not to potentially kill this project, but did you happen to see this post and the reference to forthcoming native backup workflows? https://forums.unraid.net/topic/137710-plugin-appdatabackup/page/77/#findComment-1564584

The "tool tip" pop-up on the "Update" button could use better wording, IMO. Also, when clicking Update I'm getting an error pop-up from the browser. JavaScript from "https://<ip address>" invalid cron expression But I haven't updated to your most recent version yet, as I'm in the middle of a test backup.

Yeah, I realize it takes a while via network share. This might push me to keep it local and then use something else to sync to remote. I already have Cloudberry Backup running so that could do it. It just doesn't like the VMs directly.

Must have been something else--it seems to be working fine for me as well--albeit quite slow at the moment. Another feature idea: a button to STOP/CANCEL an in-progress backup....

I believe that one needs to use Unassigned Devices to mount remote shares. Further, turning on the "share" toggle allowed the back up to complete to the remote share. UPDATE: Well, turning off that "SHARE" toggle seems to still allow it to work fine. I'm not sure if that changed something or it failed again earlier for some other reason. ¯\_(ツ)_/¯

I'm not entirely following this...it is above my pay grade...but I think, after reading some of the associated documentation, that I need to turn on the "share" toggle in the NFS Share settings. Yes?

First, you might have to clarify what you mean by "exclusive" as this is not a term/setting I have previously encountered in my Unraid meanderings. Secondly, screenshot of what? The plugin settings or the share/mount for remotes?

Another small thing. The indicated backup actually had errors (failed completely because of the mount points) upon finish. The notification in the Unraid notifications should ideally indicate that it failed, not just finished, and have the warning icon, similar to the other clip below.

I am not afraid of terminal... :-D

A few other notes: This is a jumble, particularly if you have multiple VMs selected for backup. When you click EDIT here, the EDIT button at 3 appears. This EDIT button, when you click EDIT at 2, seems awkward. Perhaps this should be "SAVE" or "UPDATE"? Love where this is going.

Updated, still seeing the same mount type error. Backup session finished - 2026-02-24 22:04:24 Backup duration: 8s ... [ERROR] Skipping PMOX BU Serv due to incompatible mount types [ERROR] They must be on the same mount type i.e both fields using user or both user0 or none using either user or user0 [ERROR] Vdisk /mnt/user/domains/PMOX BU Serv/vdisk1.img is using mount type (USER) and backup destination (OTHER) PMOX BU Serv is now stopped Stopping PMOX BU Serv Started backup for PMOX BU Serv Backing up VM(s) - PMOX BU Serv Backup session started - 2026-02-24 22:04:15 --------------------------------------------------------------------------------------------------

I mean, I guess one could use all sorts of other actions to copy/sync backups the remote shares, but personally I would prefer to limit the number of links in the chain. So this one change would be ideal, IMO.

gesundheit* *okay, i think

On a related topic, why will it not allow me to select /mnt/user/bu-cor-vms, or any folder directly under /mnt/user but it will allow me to select a folder within /mnt/user/bu-cor-vms?

I have the same issue. I would like to run backups to another device on the network. But, it is at the following /mnt/remotes/NAS_backups. So are you suggesting that this needs to be /mnt/user/NAS_backups ? I guess that works, but changing it now would break all sorts of other stuff for me.

Cycle VM: Is there any plan to allow for multiple VMs or Containers to be managed? For example: Container 1, VM-A, VM-B all need GPU. Container-1 runs normally (ollama) and should turn off when either VM-A or VM-B are started.

Cycle VM: Could this be extended to cycle docker containers that use the GPU? For example, I usually have an AI running that uses GPU, but if I forget to turn it off and I start a VM that also uses the GPU, system lockup.