搭了个三节点的Kafka集群，只要一创建topic就报Should not set log start offset的错

        at java.lang.Thread.run(Thread.java:748)
[2019-12-30 16:08:37,947] INFO [ReplicaFetcher replicaId=1, leaderId=3, fetcherId=0] Node 3 was unable to process the fetch request with (sessionId=INVALID, epoch=INITIAL): UNKNOWN_SERVER_ERROR. (org.apache.kafka.clients.FetchSessionHandler)
[2019-12-30 16:08:37,948] ERROR [KafkaApi-1] Error when handling request: clientId=broker-1-fetcher-0, correlationId=2188, api=FETCH, body={replica_id=1,max_wait_time=500,min_bytes=1,max_bytes=10485760,isolation_level=0,session_id=0,session_epoch=0,topics=[{topic=my-replicated-topic,partitions=[{partition=0,current_leader_epoch=0,fetch_offset=0,log_start_offset=0,partition_max_bytes=1048576}]}],forgotten_topics_data=[],rack_id=} (kafka.server.KafkaApis)
org.apache.kafka.common.KafkaException: Should not set log start offset on partition my-replicated-topic-0's local replica 1 without attempting to delete records of the log
        at kafka.cluster.Replica.logStartOffset_$eq(Replica.scala:142)
        at kafka.cluster.Replica.updateLogReadResult(Replica.scala:78)
        at kafka.cluster.Partition.updateReplicaLogReadResult(Partition.scala:487)
        at kafka.server.ReplicaManager.$anonfun$updateFollowerLogReadResults$2(ReplicaManager.scala:1406)
        at scala.collection.TraversableLike.$anonfun$map$1(TraversableLike.scala:237)
        at scala.collection.mutable.ResizableArray.foreach(ResizableArray.scala:62)
        at scala.collection.mutable.ResizableArray.foreach$(ResizableArray.scala:55)
        at scala.collection.mutable.ArrayBuffer.foreach(ArrayBuffer.scala:49)
        at scala.collection.TraversableLike.map(TraversableLike.scala:237)
        at scala.collection.TraversableLike.map$(TraversableLike.scala:230)
        at scala.collection.AbstractTraversable.map(Traversable.scala:108)
        at kafka.server.ReplicaManager.updateFollowerLogReadResults(ReplicaManager.scala:1400)
        at kafka.server.ReplicaManager.readFromLog$1(ReplicaManager.scala:853)
        at kafka.server.ReplicaManager.fetchMessages(ReplicaManager.scala:857)
        at kafka.server.KafkaApis.handleFetchRequest(KafkaApis.scala:757)
        at kafka.server.KafkaApis.handle(KafkaApis.scala:114)
        at kafka.server.KafkaRequestHandler.run(KafkaRequestHandler.scala:69)
        at java.lang.Thread.run(Thread.java:748)

一头雾水，百度不出解决方案，google上不了。