Be aware that we have added additional logging and retry logging in the 1.3.X versions of the client. The retry logic is for handling scenarios when the cluster is transitioning between states, for instance during a rebalance or when a node is added or removed. It's a Warn, because in subsequent retries it's likely that an active node will be selected and the operation will succeed.
Is this the error that your getting from your OperationResult.Message property or what you are seeing in your logs? The most important result is the final outcome of the operation and what you would see at your at application level and not in your logs.
That being said, how often is this happening? What is the state of the cluster when this is happening? What about at the app server level - is IIS doing an app-pool refresh at this time?
If you can supply answers to those questions and logs, it would be helpful in diagnosing what is going on here.