返回到文章

采纳

编辑于
wlh

kafka reassign partition后, ISR 落后于 Replicas

kafka

执行 reassign partition命令:

/opt/kafka/bin/kafka-reassign-partitions.sh --zookeeper 10.117.10.194:2181,10.252.223.221:2181,10.117.3.243:2181 --reassignment-json-file p4.json --execute

之后topic状态不正常

[root@prod-mq1 bin]# ./kafka-topics.sh --describe --zookeeper 10.117.10.194:2181,10.252.223.221:2181,10.117.3.243:2181 --topic ods_prod11_5_7
Topic:ods_prod11_5_7    PartitionCount:10       ReplicationFactor:3     Configs:
        Topic: ods_prod11_5_7   Partition: 0    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 1    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 2    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 3    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 4    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 5    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 6    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 7    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 8    Leader: 1       Replicas: 1,2,3 Isr: 1
        Topic: ods_prod11_5_7   Partition: 9    Leader: 1       Replicas: 1,2,3 Isr: 1

Topic 处于 under replicated 状态.

server.log 充满:

  [2018-08-08 16:21:12,773] WARN Received a PartitionLeaderEpoch assignment for an epoch < latestEpoch. This implies messages have arrived out of order. New: {epoch:13, offset:89409682}, Current: {epoch:38, offset84236994} for Partition: prod-rm-bp1p125xw0bvcey34-2 (kafka.server.epoch.LeaderEpochFileCache)

求教大家,如何让topic恢复正常?