Merge pull request #451 from hrishi18pathak/SEClusterConfigBugFix

Surface better error instead of MOVED errors during a network disconnect in redis clusters 

StackExchange.Redis/StackExchange/Redis/ResultProcessor.cs

@@ -141,6 +141,7 @@ public virtual bool SetResult(PhysicalConnection connection, Message message, Ra
                 var server = bridge.ServerEndPoint;
                 bool log = !message.IsInternalCall;
                 bool isMoved = result.AssertStarts(MOVED);
+                string err = string.Empty;
                 if (isMoved || result.AssertStarts(ASK))
                 {
                     message.SetResponseReceived();
@@ -161,11 +162,19 @@ public virtual bool SetResult(PhysicalConnection connection, Message message, Ra
                                 connection.Multiplexer.Trace(message.Command + " re-issued to " + endpoint, isMoved ? "MOVED" : "ASK");
                                 return false;
                             }
+                            else
+                            {
+                                err = string.Format("Endpoint {0} serving hashslot {1} is not reachable at this point of time. Please check connectTimeout value. If it is low, try increasing it to give the ConnectionMultiplexer a chance to recover from the network disconnect.", endpoint, hashSlot);
+                            }
                         }
                     }
                 }
 
-                string err = result.GetString();
+                if (string.IsNullOrWhiteSpace(err))
+                {
+                    err = result.GetString();
+                }
+                
                 if (log)
                 {
                     bridge.Multiplexer.OnErrorMessage(server.EndPoint, err);