Handle Tesla Fleet Authentication Failures

[BJReplay]

Slightly OT: If you've got a Tesla Powerwall and are struggling with Fleet API authentication failures (there's a fix coded, but not yet available in Home Assistant), the automation below is a temporary workaround.

The underlying problem is that there is a timing issue with authentication caused by flaky responses from the Tesla servers that falls through as a failure, when it could be handled by the integration by using the refresh token.

Restarting Home Assistant or reloading the Tesla Fleet integration both force the integration to use the refresh token, and most of the time this fixes the authentication problem.

This automation automates the reload for you, so you won't even notice that it is required.

However, for you to set it up, you'll need to wait until you've seen the repair issue raised, so that you can look at the issue ID in developer tools as you'll need that to customise the automation.

You'll also need to select your Fleet Integration to finish the customisation.

Then, run the automation, and the repair should close.

Firstly, you almost certainly need Spook installed from HACS and the integration set up:

To find out if you do, go into Developer Tools States: and search for the following entities

event.repair
sensor.active_issues

If you have them, you won't need spook, but if you don't, you will.

If you need spook, grab it:

It's possible that since I installed it, the sensors that it provides that this relies on have now been included in Home Assistant via Labs, but I believe that Spook is required.

Once you've got spook, or confirmed you've got the entities, you need to find your issue_id for the Tesla Fleet reauth repair.

Here's mine (partly redacted):

You'll need that issue_id to customise the automation.

You should be able to find it if it was the last repair - even if you've already run the repair.

Here's the automation in yaml:

alias: Repair Tesla Fleet Auth Issue
description: Repair Tesla Fleet Auth Issue
triggers:
  - trigger: numeric_state
    entity_id:
      - sensor.active_issues
    above: 0
conditions:
  - condition: state
    entity_id: event.repair
    state:
      - config_entry_reauth_tesla_fleet_REDACTED_ISSUE_ID
    attribute: issue_id
  - condition: state
    entity_id: event.repair
    state:
      - create
      - update
    attribute: event_type
actions:
  - action: homeassistant.reload_config_entry
    metadata: {}
    data:
      entry_id: REDACTED_CONFIG_ENTRY_ID
mode: single

And here's what it looks like:

You'll need to select your Tesla Fleet Integration from the drop down to complete the automation.

Once set up, when a repair is created or updated, if it is for the Tesla Fleet Reauth problem, this automation will automatically reload the Fleet API integration, which should cause the integration to use its refresh token to handle the reauth for you.

I set mine up yesterday, ran it manually when the repair was already there, and it fixed it, and it triggered automatically this morning, confirming that the trigger logic works.

[autoSteve]

This is a present thing for me. Thanks, BJ. This has been a thing in the past, IIRC.

I love that integration, and grumble at it in transient moments like this; it is quiet as a mouse until it hollers and breaks shit.

[BJReplay]

Fix has been pulled to HA Dev, but didn't make core 2026.3.4