Our cluster kafka for HDP, version is Kafka_2. 10-0.9.0.2.4.0.0-169. The jar, the spark for the native 2.3.0, when using the spark - streaming - kafka - 0-10 _2. 11 will automatically download a kafka - client 0.10.0.1 jars, such consumption is less than the datagram will lead to a wrong , excuse me this kind of circumstance have any good solution,

CodePudding user response:

Can only see your brokers and host not parsed and don't see your code is not good to conclusions