Automations stop on error - Add Option?


I noticed that when there is an error in an automation it is the whole automation that no longer works, the following actions are not executed.

I have done research and have found that several people are in this situation. Indeed, I could see various tricks with scripts but it is not the ideal when one has a lot of automations.

Lately, following an update of Alexa Media Player (useful for TTS) all my automations which use TTS no longer worked because the notify service was renamed with the latest version …


Is it possible to add an option like “skip the action if it is in error”? This would be ideal so as not to block the entire automation in the event of an error.

thank you in advance

I understand this is not what you wanted but in Node red you can create this by calling services in parallell instead of in series.
I have used this in some places where there could be an issue.

Could you explain what you mean by this?

I use AMP and haven’t noticed anything changed recently.

For the TSS I used until now the service:


Since the last version the service has become:


After understanding why all of my automations no longer worked, I renamed them so that it would work again.

Another similar problem, on I restarted HA and there is an automation that uses AMP immediately after restarting HA and the automation does not work either because AMP is not fully loaded (AMP entities are unavailable a few seconds / minutes after restarting HA).

thank you but I do not use Nod red, I have not yet looked into it because for the moment HA automation is suitable and evolving all the time :slight_smile:

I’ve never used the direct notify service since most of the time I want to make announcements to multiple devices so I just use the notify.alexa_media service along with the entity_id’s of each echo device.

So I guess that’s why I never noticed the change.

Thanks for that info.

You’re welcome :wink: