Conceptually related to #4443.

This would be the number one feature I would vote at this moment. It would also be nice if this flag could be configured as an option in ansible.cfg if you want it to be enabled all the time (it can be off by default).

Would this flag also mean that the handlers would be executed if the playbook fails the first time, or only on retry? I don't see why it shouldn't be possible to run handlers even if the playbook fails the first time.

This will not be a ansible.cfg setting because we don't want it to run all the time, this would defeat the point of "notify" only running things on change.

In those cases, just make something a task.

Anyway, look for this to come very soon.

The retry command will suggest adding --force-handlers, and --force-handlers is also a CLI switch you can select. Nothing will be adding this flag automatically.

@alanfairless and @kustodian this is now implemented via the "--force-handlers" parameter in ansible-playbook.

I would like a --skip-handlers, for provisioning reasons (My handlers, for example, reload services. Sometimes those services are not running yet.)

If you need to use --skip-handlers, than you are probably not using handlers the right way. Handlers are run at the end of the play, so just make sure to start the service before it reaches the end and you shouldn't have any problems.

Also handlers support when: condition, so you could set a condition for those cases when the service is still not running, even though I wouldn't recommend this.