A ：定义：所谓CommitFailedException，是指Consumer客户端在提交位移时出现了错误或异常，并且并不可恢复的严重异常。

B ：导致原因：
    （1）消费者端处理的总时间超过预设的max.poll.interval.ms参数值
    （2）出现一个Standalone Consumerd的独立消费者，配置的group.id重名冲突。

C ：解决方案：
    （1）减少单条消息处理的时间
    （2）增加Consumer端允许下游系统消费一批消息的最大时长
    （3）减少下游系统一次性消费的消息总数。
    （4）下游使用多线程加速消费

Synchronous auto-commit of offsets {=OffsetAndMetadata{offset=6236, leaderEpoch=null, metadata=''}} failed: Commit cannot be completed since the group has already rebalanced and assigned the partitions to another member. This means that the time between subsequent calls to poll() was longer than the configured max.poll.interval.ms, which typically implies that the poll loop is spending too much time message processing. You can address this either by increasing max.poll.interval.ms or by reducing the maximum size of batches returned in poll() with max.poll.records.
我们这边有几个项目，当用原生的kafka客户端经常出现这个报错，max.poll.interval.ms是默认值300000，max.poll.records.是2000，但是实际上数据很少，每条数据处理的时间也很短。heartbeat.interval.ms是2000，session.timeout.ms是12000。为什么经常出现这个错误。
重点是，另外几个项目用的是spring-kafka却重来没有出现过这样的报错，相关配置差不多，业务场景差不多。求指点？怎么样避免这样的问题

To use this mode, instead of subscribing to the topic using subscribe, you just call assign(Collection) with the full list of partitions that you want to consume.

     String topic = "foo";
     TopicPartition partition0 = new TopicPartition(topic, 0);
     TopicPartition partition1 = new TopicPartition(topic, 1);
     consumer.assign(Arrays.asList(partition0, partition1));
 
Once assigned, you can call poll in a loop, just as in the preceding examples to consume records. The group that the consumer specifies is still used for committing offsets, but now the set of partitions will only change with another call to assign. Manual partition assignment does not use group coordination, so consumer failures will not cause assigned partitions to be rebalanced. Each consumer acts independently even if it shares a groupId with another consumer. To avoid offset commit conflicts, you should usually ensure that the groupId is unique for each consumer instance.

老师 standalone mode 是上面这段内容吗？

老师你好，我在使用spark streaming消费kafka消息会遇到某个batch消费耗时长。请老师帮忙分析一下这个是什么原因，应该怎么去优化。谢谢老师。日志大体如下：
20/02/06 21:13:18 INFO KafkaRDD: Computing topic test_topic, partition 0 offsets 470985316 -> 470988502
20/02/06 21:13:18 DEBUG NetworkClient: Disconnecting from node 1 due to request timeout.
20/02/06 21:13:18 DEBUG ConsumerNetworkClient: Cancelled FETCH request ClientRequest(expectResponse=true, callback=org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient$RequestFutureCompletionHandler@bf99562, request=RequestSend(header={api_key=1,api_version=2,correlation_id=32,client_id=consumer-2}, body={replica_id=-1,max_wait_time=3000,min_bytes=1,topics=[{topic=test_topic,partitions=[{partition=0,fetch_offset=470986525,max_bytes=10485760}]}]}), createdTimeMs=1580994741183, sendTimeMs=1580994741183) with correlation id 32 due to node 1 being disconnected
20/02/06 21:13:18 DEBUG Fetcher: Fetch failed
org.apache.kafka.common.errors.DisconnectException
20/02/06 21:13:18 DEBUG NetworkClient: Initialize connection to node 2 for sending metadata request
20/02/06 21:13:18 DEBUG NetworkClient: Initiating connection to node 2 at host1:9092.

2019-12-23 16:43:56,367 consumer.py[line:792] WARNING Auto offset commit failed for group aff74e1e254e11ea9f47b827eb16d0ae: NodeNotReadyError: coordinator-1
2019-12-23 16:43:56,471 client_async.py[line:695] WARNING <BrokerConnection node_id=coordinator-1 host=39.104.137.50:9093 <connected> [IPv4 ('39.104.137.50', 9093)]> timed out after 305000 ms. Closing connection.
2019-12-23 16:43:56,473 client_async.py[line:327] WARNING Node coordinator-1 connection failed -- refreshing metadata
2019-12-23 16:43:56,478 base.py[line:493] ERROR Error sending HeartbeatRequest_v1 to node coordinator-1 [[Error 7] RequestTimedOutError: Request timed out after 305000 ms]
2019-12-23 16:43:56,479 base.py[line:714] WARNING Marking the coordinator dead (node coordinator-1) for group aff74e1e254e11ea9f47b827eb16d0ae: [Error 7] RequestTimedOutError: Request timed out after 305000 ms.
出现上面的报错consumer就卡主了，不能接收数据了，是提交失败吧

[2019-07-17 18:53:36,230] ERROR [ReplicaManager broker=2] Error processing append operation on partition __consumer_offsets-49 (kafka.server.ReplicaManager)
org.apache.kafka.common.errors.NotEnoughReplicasException: The size of the current ISR Set(2) is insufficient to satisfy the min.isr requirement of 2 for partition __consumer_offsets-49
把min.insync.replicas = 2改成1，消费者就可以运行了，这种flower失效的情况怎么处理呢？