If driver enters failure-recovery loop, this should be detected and promoted to an error.
A flexible policy should be implemented with meaningful default.
Expose failures and recoveries as IObservable and subscribe a policy which would raise an error when failure-recovery pattern happens for more than 1000 times within 5 minutes and no message has been received (no progress).
If driver enters failure-recovery loop, this should be detected and promoted to an error. A flexible policy should be implemented with meaningful default.
Expose failures and recoveries as IObservable and subscribe a policy which would raise an error when failure-recovery pattern happens for more than 1000 times within 5 minutes and no message has been received (no progress).