【kafka】__consumer_offsets部分分区异常导致消费不到数据问题排查

Error while executing consumer group command The group coordinator is not available.
org.apache.kafka.common.errors.GroupCoordinatorNotAvailableException: The group coordinator is not available.

19:52:41.785 TKD [main] DEBUG o.a.k.c.c.i.AbstractCoordinator - Issuing group metadata request to broker 43
19:52:41.788 TKD [main] DEBUG o.a.k.c.c.i.AbstractCoordinator - Group metadata response ClientResponse(receivedTimeMs=1587642761788, disconnected=false, request=ClientRequest(expectResponse=true, callback=org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient$RequestFutureCompletionHandler@1b68ddbd, request=RequestSend(header={api_key=10,api_version=0,correlation_id=30,client_id=consumer-1}, body={group_id=30cab231-05ed-43ef-96aa-a3ca1564baa3}), createdTimeMs=1587642761785, sendTimeMs=1587642761785), responseBody={error_code=15,coordinator={node_id=-1,host=,port=-1}})
19:52:41.875 TKD [main] DEBUG o.apache.kafka.clients.NetworkClient - Sending metadata request ClientRequest(expectResponse=true, callback=null, request=RequestSend(header={api_key=3,api_version=0,correlation_id=31,client_id=consumer-1}, body={topics=[topic打马赛克]}), isInitiatedByNetworkClient, createdTimeMs=1587642761875, sendTimeMs=0) to node 43

  def handle(request: RequestChannel.Request) {
  ……
        case RequestKeys.GroupCoordinatorKey => handleGroupCoordinatorRequest(request)
  ……
    }

  def handleGroupCoordinatorRequest(request: RequestChannel.Request) {
    val groupCoordinatorRequest = request.body.asInstanceOf[GroupCoordinatorRequest]
    val responseHeader = new ResponseHeader(request.header.correlationId)

    if (!authorize(request.session, Describe, new Resource(Group, groupCoordinatorRequest.groupId))) {
      val responseBody = new GroupCoordinatorResponse(Errors.GROUP_AUTHORIZATION_FAILED.code, Node.noNode)
      requestChannel.sendResponse(new RequestChannel.Response(request, new ResponseSend(request.connectionId, responseHeader, responseBody)))
    } else {
      val partition = coordinator.partitionFor(groupCoordinatorRequest.groupId)

      // get metadata (and create the topic if necessary)
      val offsetsTopicMetadata = getOrCreateGroupMetadataTopic(request.securityProtocol)
        // 第一个可能存在的问题:offsetsTopicMetadata的errCode不为空
      val responseBody = if (offsetsTopicMetadata.errorCode != Errors.NONE.code) {
        new GroupCoordinatorResponse(Errors.GROUP_COORDINATOR_NOT_AVAILABLE.code, Node.noNode)
      } else {
        val coordinatorEndpoint = offsetsTopicMetadata.partitionsMetadata
          .find(_.partitionId == partition)
          .flatMap {
            partitionMetadata => partitionMetadata.leader
          }
        // 第二个可能存在的问题:coordinatorEndpoint为空
        coordinatorEndpoint match {
          case Some(endpoint) =>
            new GroupCoordinatorResponse(Errors.NONE.code, new Node(endpoint.id, endpoint.host, endpoint.port))
          case _ =>
            new GroupCoordinatorResponse(Errors.GROUP_COORDINATOR_NOT_AVAILABLE.code, Node.noNode)
        }
      }

      trace("Sending consumer metadata %s for correlation id %d to client %s."
        .format(responseBody, request.header.correlationId, request.header.clientId))
      requestChannel.sendResponse(new RequestChannel.Response(request, new ResponseSend(request.connectionId, responseHeader, responseBody)))
    }
  }

  def partitionFor(groupId: String): Int = Utils.abs(groupId.hashCode) % groupMetadataTopicPartitionCount

scala> "group1-打马赛克".hashCode % 50
res2: Int = 43

scala> "group2-打马赛克".hashCode % 50
res3: Int = 43

val offsetsTopicMetadata = getOrCreateGroupMetadataTopic(request.securityProtocol)

  private def getPartitionMetadata(topic: String, protocol: SecurityProtocol): Option[Iterable[PartitionMetadata]] = {
    cache.get(topic).map { partitions =>
      partitions.map { case (partitionId, partitionState) =>
        val topicPartition = TopicAndPartition(topic, partitionId)

        val leaderAndIsr = partitionState.leaderIsrAndControllerEpoch.leaderAndIsr
        val maybeLeader = aliveBrokers.get(leaderAndIsr.leader)

        val replicas = partitionState.allReplicas
        val replicaInfo = getAliveEndpoints(replicas, protocol)

        maybeLeader match {
          case None =>
            debug("Error while fetching metadata for %s: leader not available".format(topicPartition))
            new PartitionMetadata(partitionId, None, replicaInfo, Seq.empty[BrokerEndPoint],
              Errors.LEADER_NOT_AVAILABLE.code)

          case Some(leader) =>
            val isr = leaderAndIsr.isr
            val isrInfo = getAliveEndpoints(isr, protocol)

            if (replicaInfo.size < replicas.size) {
              debug("Error while fetching metadata for %s: replica information not available for following brokers %s"

  }