HI,
I’m new. Hopefully I’m in the right place. I’m trying to update the poller so that it can handle long running jobs over flaky connections.
So far, I’ve found that the playbook class method _async_poll which calls poller.wait. Seems like this it the area I want to make changes in, but still lost as to how to approach this. Any hints would be helpful. Thanks!
-Dan
I think the better way to handle connection retries would be at the task level independent of async.
We are probably going to have this a lot easier to add once the ‘v2/’ code tree switches over.
I’d hold up on pull requests in this area right now as such a thing really shouldn’t be specific to an async task and should probably be considered a play level directive (at the same level as “hosts:”).
Sort of make sense?