Blog
flink问题集
1、flink-sql中消费kafka策略group-offsets
CREATE TABLE t2 (
id BIGINT,
ts BIGINT,
vc BIGINT
) WITH (
'connector' = 'kafka',
'topic' = 't2',
'properties.bootstrap.servers' = 'hadoop001:9092',
'properties.group.id' = 'test-t2',
'scan.startup.mode' = 'group-offsets',
'format' = 'json'
);
group-offets 应该是指定分组的latest策略,最终执行的是setStartingOffsets(OffsetsInitializer.committedOffsets(OffsetResetStrategy.LATEST))