错误日志：

2024-01-08 10:45:44
org.apache.flink.util.FlinkException: Global failure triggered by OperatorCoordinator for 'Source: Kafka Source 主流' (operator bc764cd8ddf7a0cff126f51c16239658).
	at org.apache.flink.runtime.operators.coordination.OperatorCoordinatorHolder$LazyInitializedCoordinatorContext.failJob(OperatorCoordinatorHolder.java:545)
	at org.apache.flink.runtime.operators.coordination.RecreateOnResetOperatorCoordinator$QuiesceableContext.failJob(RecreateOnResetOperatorCoordinator.java:223)
	at org.apache.flink.runtime.source.coordinator.SourceCoordinatorContext.failJob(SourceCoordinatorContext.java:285)
	at org.apache.flink.runtime.source.coordinator.SourceCoordinatorContext.handleUncaughtExceptionFromAsyncCall(SourceCoordinatorContext.java:298)
	at org.apache.flink.util.ThrowableCatchingRunnable.run(ThrowableCatchingRunnable.java:42)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
Caused by: org.apache.flink.util.FlinkRuntimeException: Failed to list subscribed topic partitions due to 
	at org.apache.flink.connector.kafka.source.enumerator.KafkaSourceEnumerator.checkPartitionChanges(KafkaSourceEnumerator.java:237)
	at org.apache.flink.runtime.source.coordinator.ExecutorNotifier.lambda$null$1(ExecutorNotifier.java:86)
	at org.apache.flink.util.ThrowableCatchingRunnable.run(ThrowableCatchingRunnable.java:40)
	... 3 more
Caused by: java.lang.RuntimeException: Failed to get metadata for topics [pre_topic_cunsumer_test].
	at org.apache.flink.connector.kafka.source.enumerator.subscriber.KafkaSubscriberUtils.getTopicMetadata(KafkaSubscriberUtils.java:47)
	at org.apache.flink.connector.kafka.source.enumerator.subscriber.TopicListSubscriber.getSubscribedTopicPartitions(TopicListSubscriber.java:52)
	at org.apache.flink.connector.kafka.source.enumerator.KafkaSourceEnumerator.getSubscribedTopicPartitions(KafkaSourceEnumerator.java:222)
	at org.apache.flink.runtime.source.coordinator.ExecutorNotifier.lambda$notifyReadyAsync$2(ExecutorNotifier.java:83)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
	... 3 more
Caused by: java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.TimeoutException: Timed out waiting for a node assignment.
	at org.apache.kafka.common.internals.KafkaFutureImpl.wrapAndThrow(KafkaFutureImpl.java:45)
	at org.apache.kafka.common.internals.KafkaFutureImpl.access$000(KafkaFutureImpl.java:32)
	at org.apache.kafka.common.internals.KafkaFutureImpl$SingleWaiter.await(KafkaFutureImpl.java:89)
	at org.apache.kafka.common.internals.KafkaFutureImpl.get(KafkaFutureImpl.java:260)
	at org.apache.flink.connector.kafka.source.enumerator.subscriber.KafkaSubscriberUtils.getTopicMetadata(KafkaSubscriberUtils.java:44)
	... 10 more
Caused by: org.apache.kafka.common.errors.TimeoutException: Timed out waiting for a node assignment.
2024-01-08 10:45:44
org.apache.flink.util.FlinkException: Global failure triggered by OperatorCoordinator for 'Source: Kafka Source 主流' (operator bc764cd8ddf7a0cff126f51c16239658).
	at org.apache.flink.runtime.operators.coordination.OperatorCoordinatorHolder$LazyInitializedCoordinatorContext.failJob(OperatorCoordinatorHolder.java:545)
	at org.apache.flink.runtime.operators.coordination.RecreateOnResetOperatorCoordinator$QuiesceableContext.failJob(RecreateOnResetOperatorCoordinator.java:223)
	at org.apache.flink.runtime.source.coordinator.SourceCoordinatorContext.failJob(SourceCoordinatorContext.java:285)
	at org.apache.flink.runtime.source.coordinator.SourceCoordinatorContext.handleUncaughtExceptionFromAsyncCall(SourceCoordinatorContext.java:298)
	at org.apache.flink.util.ThrowableCatchingRunnable.run(ThrowableCatchingRunnable.java:42)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
Caused by: org.apache.flink.util.FlinkRuntimeException: Failed to list subscribed topic partitions due to 
	at org.apache.flink.connector.kafka.source.enumerator.KafkaSourceEnumerator.checkPartitionChanges(KafkaSourceEnumerator.java:237)
	at org.apache.flink.runtime.source.coordinator.ExecutorNotifier.lambda$null$1(ExecutorNotifier.java:86)
	at org.apache.flink.util.ThrowableCatchingRunnable.run(ThrowableCatchingRunnable.java:40)
	... 3 more
Caused by: java.lang.RuntimeException: Failed to get metadata for topics [pre_topic_cunsumer_test].
	at org.apache.flink.connector.kafka.source.enumerator.subscriber.KafkaSubscriberUtils.getTopicMetadata(KafkaSubscriberUtils.java:47)
	at org.apache.flink.connector.kafka.source.enumerator.subscriber.TopicListSubscriber.getSubscribedTopicPartitions(TopicListSubscriber.java:52)
	at org.apache.flink.connector.kafka.source.enumerator.KafkaSourceEnumerator.getSubscribedTopicPartitions(KafkaSourceEnumerator.java:222)
	at org.apache.flink.runtime.source.coordinator.ExecutorNotifier.lambda$notifyReadyAsync$2(ExecutorNotifier.java:83)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
	... 3 more
Caused by: java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.TimeoutException: Timed out waiting for a node assignment.
	at org.apache.kafka.common.internals.KafkaFutureImpl.wrapAndThrow(KafkaFutureImpl.java:45)
	at org.apache.kafka.common.internals.KafkaFutureImpl.access$000(KafkaFutureImpl.java:32)
	at org.apache.kafka.common.internals.KafkaFutureImpl$SingleWaiter.await(KafkaFutureImpl.java:89)
	at org.apache.kafka.common.internals.KafkaFutureImpl.get(KafkaFutureImpl.java:260)
	at org.apache.flink.connector.kafka.source.enumerator.subscriber.KafkaSubscriberUtils.getTopicMetadata(KafkaSubscriberUtils.java:44)
	... 10 more
Caused by: org.apache.kafka.common.errors.TimeoutException: Timed out waiting for a node assignment.

解决方案：

多台机器都要配置下/etc/hosts的配置映射

排查过程：

先看了下kafka的进程信息

ps -ef |grep kafka

然后看了下如下配置信息

/var/run/cloudera-scm-agent/process/405-kafka-KAFKA_BROKER/kafka.properties

这里看到里面的zk使用了映射的名字

猜测因为zk的连接过程是通过映射名称来通信的