Flink SQL -- CheckPoint

1、开启CheckPoint

checkpoint可以定时将flink任务的状态持久化到hdfs中，任务执行失败重启可以保证中间结果不丢失

# 修改flink配置文件
vim flink-conf.yaml

# checkppint 间隔时间
execution.checkpointing.interval: 1min
# 任务手动取消时保存checkpoint
execution.checkpointing.externalized-checkpoint-retention: RETAIN_ON_CANCELLATION
# 同时允许1个checkpoint执行
execution.checkpointing.max-concurrent-checkpoints: 1
execution.checkpointing.min-pause: 0
# 数据处理的语义
execution.checkpointing.mode: EXACTLY_ONCE
# checkpoint超时时间
execution.checkpointing.timeout: 10min
execution.checkpointing.tolerable-failed-checkpoints: 0
execution.checkpointing.unaligned: false
# 状态后端（保存状态的位置，hashmap：内存）
state.backend: hashmap
# checkpoint路径
state.checkpoints.dir: hdfs://master:9000/flink/checkpoint

2、编写一个Flnik SQL 脚本：

vim word_count.sql

-- 实时从kafka中读取单词，统计单词的数量，将结果保存到mysql中

-- 1、创