A lot of log files might be pulled on errors

[Fobhep]

First up: The new "get log back to ansible controller" feature is really awesome - thanks a lot :)

However one might end up with quite a bunch of logs on the controller. Do you think a possible feature would be to record the timestamp of the deployment start and pull only logs that are younger than that? Maybe that could be called slim_logging :)

Also eg I am not sure if I really need all types of kafka logs (requests, authorizer etc) for debugging a failed deployment

[domenicbove]

Yes this is a great thought, I recently was debugging something w a user and we had to remove some kafka log files because he could not email the entire tar. At least we tar it haha. I think the GC logs aren't helpful/ waste space.

Do you have the full contents of one of these tars that you can paste? Lets try and pin point the "non-useful" files!

[Fobhep]

sure - I will get back at you asap with that

[utkarsh5474]

Hey @Fobhep We would love to get such a list if you get time.