I am thinking of adding support to retry failed hosts in ansible-playbook cli. Is there any reason why we should not support that?
Huy
We have limited retry with do/until but I could see the usefulness for
those with 'problem hosts' of a 'retries: #' entry at play/task/global
levels.
+1 .
I had to work with such a system where I had to retry everywhere.
It’s also handy for another problem I encountered in the past, I wrote a set of Ansible playbooks to interact with a web service where everything is asynchronous and requires status polling everywhere. Having a global retry would’ve simplified those playbooks.