Can't mount NFS, please help

Installation
1

I can’t mount my NFS. It used to work, I updated, now it won’t reconnect and I can’t add network storage.
I get Transaction for mnt-data-supervisor-mounts-backups.mount/stop is destructive (mnt-data-supervisor-mounts-backups.mount has ‘start’ job queued, but ‘stop’ is included in transaction).
Rebooting didn’t help.

It’s an HA Green

  • Installation method Home Assistant OS
  • Core 2026.2.0
  • Supervisor 2026.01.1
  • Operating System 17.0
  • Frontend 20260128.6
2

I tried restoring a backup from when it last worked. That did not fix NFS at all.

3

I was struggling with this same issue this morning and found a solution that worked for me. I wanted to share in case it helps anyone else.

In this case, I was trying to remount my NFS share named “frigate”, you should replace <name> with backups

Enable Home Assistant OS Host SSH (port 22222)

  1. Format a USB drive with a partition named CONFIG (all caps, case sensitive) as FAT32
  2. Generate an SSH key on the machine you’ll connect from (skip if you already have one):
   ssh-keygen -t ed25519
  1. Copy your public key to a file called authorized_keys (no extension) in the root of the USB drive:
   cat ~/.ssh/id_ed25519.pub > <path-to-CONFIG>/authorized_keys
  1. Plug the USB into your Home Assistant machine and reboot the host machine, not just Home Assistant. Reboot through Settings > System > Power Button in top-left > Advanced Options. Or in the core terminal:
   ha host reboot

  1. Import the keys using the with core terminal:
    ha os import

  2. Connect to the OS through SSH:

   ssh root@<ha-ip> -p 22222

Reset the Supervisor mount units and jobs using OS SSH

  1. List Supervisor-managed mount units and see if there are any failed units.
    systemctl list-units --type=mount | grep <name>

  2. Reset failed systemd units

systemctl reset-failed mnt-data-supervisor-mounts-<name>.mount
systemctl daemon-reexec

  1. Check for queued systemd jobs
    systemctl list-jobs | grep <name>

  2. Cancel any stuck start/stop jobs that are related to mounts
    systemctl cancel <job-id>

  3. Re-add the mount with UI like normal, or with HA OS CLI:
    ha mounts add <name> --usage media --type nfs --server <ip> --path <path>