It'd be super handy if these also appeared in some form as Prometheus metrics, which would give us an idea of how many clients are failing to connect (ideally with a statusCode label). That way, we can alert proactively for things like version out of date errors.
It'd be super handy if these also appeared in some form as Prometheus metrics, which would give us an idea of how many clients are failing to connect (ideally with a statusCode label). That way, we can alert proactively for things like version out of date errors.