ChangbingChen opened a new issue #4618:
URL: https://github.com/apache/hudi/issues/4618
**Describe the problem you faced**
When querying a hudi table in hive, there have duplated records.
This hudi table is produced by flink.
**To Reproduce**
Steps to reproduce the behavior:
1. submit a flink job
flink-sql-client -f mysql_table_sink.sql
the sql file content:
create table `mysql_table_kafka` (
`id` bigint,
`create_time` timestamp,
`update_time` timestamp
)
with (
'connector' = 'kafka',
'topic' = 'mysql_table_kafka',
'properties.bootstrap.servers' = 'x.x.x.x:9092,x.x.x.x:9092,x.x.x.x:9092',
'properties.group.id' = 'mysql_table_cg',
'format' = 'canal-json',
'scan.startup.mode' = 'latest-offset'
);
[INFO] Execute statement succeed.
Flink SQL>
create table `mysql_table_sink_new` (
`id` bigint,
`create_time` bigint,
`update_time` bigint
`dt` varchar(20)
) partitioned by (`dt`)
with(
'connector' = 'hudi'
,'path' = 'hdfs://nameservice1/hudi/mysql_table_sink_new'
,'hoodie.datasource.write.recordkey.field' = 'id'
,'write.precombine.field' = 'create_time'
,'write.tasks' = '1'
,'compaction.tasks' = '1'
,'read.streaming.enabled' = 'true'
,'read.streaming.check-interval' = '60'
,'table.type' = 'MERGE_ON_READ'
,'compaction.async.enabled' = 'true'
,'compaction.trigger.strategy' = 'num_commits'
,'compaction.delta_commits' = '2'
,'hive_sync.enable' = 'true'
,'hive_sync.mode' = 'hms'
,'hive_sync.metastore.uris' = 'thrift://x.x.x.x:9083'
,'hive_sync.jdbc_url' = 'jdbc:hive2://x.x.x.x:10000'
,'hive_sync.table' = 'mysql_table_sink_new'
,'hive_sync.db' = 'hv_ods'
);
insert into mysql_table_sink_new select
id
,unix_timestamp(date_format(create_time, 'yyyy-MM-dd HH:mm:ss'))*1000
,unix_timestamp(date_format(update_time, 'yyyy-MM-dd HH:mm:ss'))*1000
,date_format(create_time, 'yyyyMMdd') as dt from mysql_table_kafka;
2.query in beeline
select
m
,sum(case when cnt=1 then 1 else 0 end) as one_cnt
,sum(case when cnt=2 then 1 else 0 end) as two_cnt
,sum(case when cnt=3 then 1 else 0 end) as three_cnt
,sum(case when cnt=4 then 1 else 0 end) as four_cnt
from(
select id,from_unixtime(cast(create_time/1000 as bigint), 'yyyy-MM-dd
HH:mm') as m,count(1) as cnt
from mysql_table_sink_new_ro
where dt='20220117'
group by id,from_unixtime(cast(create_time/1000 as bigint), 'yyyy-MM-dd
HH:mm')
)t
group by m;
+-------------------+----------+----------+------------+-----------+--+
| m | one_cnt | two_cnt | three_cnt | four_cnt |
+-------------------+----------+----------+------------+-----------+--+
| 2022-01-17 16:07 | 0 | 0 | 0 | 5 |
| 2022-01-17 16:08 | 0 | 0 | 0 | 273 |
| 2022-01-17 16:09 | 0 | 0 | 37 | 241 |
| 2022-01-17 16:10 | 0 | 0 | 340 | 0 |
| 2022-01-17 16:11 | 0 | 21 | 239 | 0 |
| 2022-01-17 16:12 | 0 | 253 | 0 | 0 |
| 2022-01-17 16:13 | 38 | 261 | 0 | 0 |
| 2022-01-17 16:14 | 283 | 0 | 0 | 0 |
| 2022-01-17 16:15 | 247 | 0 | 0 | 0 |
+-------------------+----------+----------+------------+-----------+--+
select id,count(1) from mysql_table_sink_new_ro group by id having
count(1)>1 limit 10;
+------------+------+--+
| id | _c1 |
+------------+------+--+
| 413588661 | 5 |
| 413588664 | 5 |
| 413588667 | 5 |
| 413588670 | 5 |
| 413588673 | 5 |
| 413588676 | 5 |
| 413588679 | 5 |
| 413588682 | 5 |
| 413588685 | 5 |
| 413588688 | 5 |
+------------+------+--+
select
`_hoodie_commit_time`,`_hoodie_commit_seqno`,`_hoodie_record_key`,`_hoodie_partition_path`,`_hoodie_file_name`,id,create_time,update_time
from mysql_table_sink_new_ro where id='413588660';
+----------------------+-----------------------+---------------------+-------------------------+----------------------------------------------------+------------+----------------+----------------+--+
| _hoodie_commit_time | _hoodie_commit_seqno | _hoodie_record_key |
_hoodie_partition_path | _hoodie_file_name |
id | create_time | update_time |
+----------------------+-----------------------+---------------------+-------------------------+----------------------------------------------------+------------+----------------+----------------+--+
| 20220117160954 | 20220117160954_0_482 | 413588661 |
20220117 |
fd45d59f-eef8-402d-a63c-6e7e5cfb5f63_0-1-0_20220117160954.parquet | 413588661
| 1642406878000 | 1642406878000 |
| 20220117160954 | 20220117160954_0_482 | 413588661 |
20220117 |
fd45d59f-eef8-402d-a63c-6e7e5cfb5f63_0-1-0_20220117160954.parquet | 413588661
| 1642406878000 | 1642406878000 |
| 20220117160954 | 20220117160954_0_482 | 413588661 |
20220117 |
fd45d59f-eef8-402d-a63c-6e7e5cfb5f63_0-1-0_20220117160954.parquet | 413588661
| 1642406878000 | 1642406878000 |
| 20220117160954 | 20220117160954_0_482 | 413588661 |
20220117 |
fd45d59f-eef8-402d-a63c-6e7e5cfb5f63_0-1-0_20220117160954.parquet | 413588661
| 1642406878000 | 1642406878000 |
| 20220117160954 | 20220117160954_0_482 | 413588661 |
20220117 |
fd45d59f-eef8-402d-a63c-6e7e5cfb5f63_0-1-0_20220117160954.parquet | 413588661
| 1642406878000 | 1642406878000 |
+----------------------+-----------------------+---------------------+-------------------------+----------------------------------------------------+------------+----------------+----------------+--+
**Expected behavior**
A clear and concise description of what you expected to happen.
**Environment Description**
* Hudi version : 0.10.0
* Spark version : xxx
* Hive version : 1.1.0-cdh5.13.3
* Hadoop version : 2.6.0-cdh5.13.3
* Storage (HDFS/S3/GCS..) : HDFS
* Running on Docker? (yes/no) : no
* Flink version : 1.13.3
**Additional context**
Add any other context about the problem here.
**Stacktrace**
```Add the stacktrace of the error.```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
