[HUDI-5058] The primary key cannot be empty when Flink reads an error from the hudi table

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.12.1
• Fix Version/s: 0.12.2, 0.13.1, 0.12.3
• Component/s: flink
• Labels: pull-request-available
• Assignee: waywtdcc
• Reporter: waywtdcc
• Created: 19/Oct/22 07:34
• Updated: 22/May/23 09:06
• Resolved: 08/Nov/22 21:57

Description

The primary key cannot be empty when Flink reads an error from the hudi table. Spark SQL is used to create tables and Spark writes data. Then the flash reports an error when reading.

spark sql write ,USE HUDI 0.11.1

 

 

create table test_hudi_cc16 (
  id bigint,
  name string,
  name2 string,
  ts bigint
) using hudi
tblproperties (
  type = 'cow',
  primaryKey = 'id' );
 insert into test_hudi_cc16
values
(1, 'cc2', 'cc32',12);
 

 

flink read, use hudi 0.12.1:

 

CREATE CATALOG myhudi WITH(
    'type' = 'hudi',
    'default-database' = 'test_hudi1',
    'catalog.path' = '/user/hdpu/warehouse',
    'mode' = 'hms',
    'hive.conf.dir' = 'hdfs:///user/hdpu/streamx/conf_data/hive_conf'
)

select *
from myhudi.test_hudi6.test_hudi_cc16; 

 

error:

 

org.apache.flink.table.api.ValidationException: Invalid primary key 'PK_id'. Column 'id' is nullable.    at org.apache.flink.table.catalog.DefaultSchemaResolver.validatePrimaryKey(DefaultSchemaResolver.java:352)
    at org.apache.flink.table.catalog.DefaultSchemaResolver.resolvePrimaryKey(DefaultSchemaResolver.java:312)
    at org.apache.flink.table.catalog.DefaultSchemaResolver.resolve(DefaultSchemaResolver.java:88)
    at org.apache.flink.table.api.Schema.resolve(Schema.java:123)
    at org.apache.flink.table.catalog.CatalogManager.resolveCatalogTable(CatalogManager.java:877)
    at org.apache.flink.table.catalog.CatalogManager.resolveCatalogBaseTable(CatalogManager.java:863)
    at org.apache.flink.table.catalog.CatalogManager.getPermanentTable(CatalogManager.java:426)
    at org.apache.flink.table.catalog.CatalogManager.getTable(CatalogManager.java:395)
    at org.apache.flink.table.api.internal.TableEnvironmentImpl.executeInternal(TableEnvironmentImpl.java:1061)
    at org.apache.flink.table.api.internal.TableEnvironmentImpl.executeSql(TableEnvironmentImpl.java:649)
    at org.grg_banking.flink.sqlexecute.FlinkUtils.exeucteSqlFile2(FlinkUtils.java:260)
    at org.apache.flink.catalog.test.TestCatalog.test1(TestCatalog.java:49)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at org.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:50)
    at org.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:12)
    at org.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:47)
    at org.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:17)
    at org.junit.internal.runners.statements.RunBefores.evaluate(RunBefores.java:26)
    at org.junit.runners.ParentRunner.runLeaf(ParentRunner.java:325)
    at org.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:78)
    at org.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:57)
    at org.junit.runners.ParentRunner$3.run(ParentRunner.java:290)
    at org.junit.runners.ParentRunner$1.schedule(ParentRunner.java:71)
    at org.junit.runners.ParentRunner.runChildren(ParentRunner.java:288)
    at org.junit.runners.ParentRunner.access$000(ParentRunner.java:58)
    at org.junit.runners.ParentRunner$2.evaluate(ParentRunner.java:268)
    at org.junit.runners.ParentRunner.run(ParentRunner.java:363)
    at org.junit.runner.JUnitCore.run(JUnitCore.java:137)
    at com.intellij.junit4.JUnit4IdeaTestRunner.startRunnerWithArgs(JUnit4IdeaTestRunner.java:69)
    at com.intellij.rt.junit.IdeaTestRunner$Repeater.startRunnerWithArgs(IdeaTestRunner.java:33)
    at com.intellij.rt.junit.JUnitStarter.prepareStreamsAndStart(JUnitStarter.java:235)
    at com.intellij.rt.junit.JUnitStarter.main(JUnitStarter.java:54)
Process finished with exit code -1
 

 

 

It may be a version compatibility problem. Flink hudi 0.12.1 does not support reading the table of spark hudi 0.11. However, I think it should be backward compatible with the read function.

 

Issue Links

• links to: GitHub Pull Request #7009
• links to: GitHub Pull Request #7981