Scheduled auto-update for Home Assistant

I think I got the same issue here… I will take a look…

1 Like

Is there an issue with updating ESPHome firmware? It hasn’t been able to update mine since ESPHome version 2024.12.4 and I believe HA’s 2025.1 update. Looking at the logbook entries for the update process, I see it attempts to, but then it doesn’t end up happening. I can’t see any failed update logs though. Manually updating an ESPHome device from HA works fine.

Loving this automation, but sadly it started failing on me today :confused:

Here’s the error:


Home Assistant Auto-update on a schedule base uses an unknown actionError ⸱ Reported by Automation

The automation “Home Assistant Auto-update on a schedule base” (automation.home_assistant_auto_update_on_a_schedule_base) has an unknown action: logbook.log.

This error prevents the automation from running correctly. Maybe this action is no longer available, or perhaps a typo caused it.

To fix this error, edit the automation and remove this action.

Select Submit below to confirm you have fixed this automation.


I tried removing the action, but it is not showing as it is a blueprint and I’m somewhat scared of touching there…!

Are you still having this issue?
My apologies, but I’ve missed your message and just saw it. I don’t have that error on my environment.

This sounds like you don’t have the logbook integration configured which is enabled by default.

Hey if I’ve defined the following:

alias: Home Assistant Auto-update
description: ""
use_blueprint:
  path: edwardtfn/auto_update_scheduled.yaml
  input:
    schedule_entity: schedule.home_assistant_auto_update_schedule
    backup_bool: false
    restart_type: core
    core_os_update_mode: all
    firmware_update_mode: all
    general_update_mode: all

Shouldn’t it update me from Home Assistant Operating System Installed version 16.0 to Latest version 16.1?

As its not doing so for me, might I be misunderstanding?

Yeah, it should be updating your OS.
Do you see anything related to this in your Logbook around the time it is scheduled to run?

Its run on a schedule yes. I’ve just triggered it now and I’m seeing this:

Logger: homeassistant.components.automation.home_assistant_auto_update
Source: helpers/script.py:2067
integration: Automation (documentation, issues)
First occurred: 13 August 2025 at 18:45:07 (5 occurrences)
Last logged: 10:37:01

Home Assistant Auto-update: Update - Standard: Error executing script. Error for call_service at pos 4: Entity update.plex_media_server_NAME does not support action update.install

K, I’ve excluded that entity and that’s fixed it :+1:

1 Like

Nice. I will look at the code anyways as it should survive to an error like this.

Hi, i used this blueprint in an automation for many months and never had any problems. But e few weeks ago it stopped working. The automation is triggered, but its not updating. I excluded non of the updates, but the list of open updates is getting longer and longer…

1 Like

Do you see anything related to this in your logs (Settings > System > Logs)?

i think not, but im not sure. can you tell me what i have to look for and where? There are differnet logs avaiable, like Host, Supervisor or Core.

Oh and btw, i did the HA Core and Operating System Updates manually as well as restarting the system. But nothing changed…

Logger: homeassistant.helpers.script
Quelle: helpers/script.py:701
Erstmals aufgetreten: 05:00:01 (1 Vorkommnis)
Zuletzt protokolliert: 05:00:01

Error in ‘if[0]’ evaluation: In ‘template’ condition: ValueError: Template error: as_timestamp got invalid input ‘None’ when rendering template ‘{{ as_timestamp(now()) - as_timestamp(last_backup_timestamp) > input_max_backup_age_seconds }}’ but no default was specified

yaml:

alias: xSystem - AutoUpdate Test
description: ""
use_blueprint:
  path: edwardtfn/auto_update_scheduled.yaml
  input:
    schedule_entity: schedule.automatisches_update_zeitfenster
    core_os_update_mode: patches
    firmware_update_mode: patches
    general_update_mode: patches
    max_backup_age:
      hours: 2
      minutes: 0
      seconds: 0
      days: 0
    restart_type: core
    verbose_logging_bool: true
    actions_pos_update:
      - action: notify.mobile_app_dergeraet
        metadata: {}
        data:
          message: Update ist gelaufen ?
          title: Update test

What can I do to fix the error? I have already re-imported the data, but unfortunately without any change.

:frowning: The problem still exists. I have to update manually. Is there any other solution?

Hi Edward,

first of all, thanks for the blueprint. I think I found a bug in the backup section that causes “unknown entity” warnings (e.g., Spook) and can also break the “max backup age” logic.

Even if a user selects a Backup state sensor via the blueprint input (Backup state sensor / run_during_backup), the blueprint still references a hard-coded entity ID:

sensor.backup_state

Example from the YAML (Backup → “Check existing backups uploaded”):
• not is_state(“sensor.backup_state”, “unknown”)
• not is_state(“sensor.backup_state”, “unavailable”)

In my setup, the backup integration provides a different entity id (e.g. sensor.backup_backup_manager_zustand), so sensor.backup_state does not exist and HA reports it as unknown.

Thanks!

1 Like

I’m in a trip right now and will look at this when back home, ok?

Of course, enjoy your time and thanks for your reply and looking into this! BR

1 Like

Hi!

i have the same issue.
It does not run. Does it run on your end and is it just a warning or is it broken?
Thank you

Hello Edward,

is there any update regarding the “sensor.backup_state” issue? I had to downgrade to v2024.9.30.0 to keep the system running.

Best regards