loveShyBoy commented on issue #5911: URL: https://github.com/apache/hudi/issues/5911#issuecomment-1161130862
> @loveShyBoy Are you using any particular [index type](https://hudi.apache.org/docs/configurations/#indextype) ? Can you share all your write configs? cc @danny0405 @yuzhaojing delta streamer used default index type , flink also used hoodie.index.type=BLOOM 0.10.1 do not have index.type config , exists since 0.11.0 `--hoodie.table.name=tablename_hudi --table.type=COPY_ON_WRITE --hoodie.datasource.write.recordkey.field=primary_id --write.precombine.field=update_time --hoodie.datasource.write.partitionpath.field=dt --hive_sync.db=ods --write.tasks=4 --hoodie.datasource.write.keygenerator.type=SIMPLE --write.bucket_assign.tasks=6 --max.poll.records=100 --max.partition.fetch.bytes=524288 --hoodie.datasource.write.hive_style_partitioning=true --index.bootstrap.enabled=false --write.index_bootstrap.tasks=6 --index.state.ttl=0.2 --prop:hoodie.index.type=BLOOM` in 0.10.1 flink used default index type is HoodieIndex.IndexType.INMEMORY，so i changed to BLOOM. but the doc say Bloom is default . I'm a little confused。that is right? ` public static final ConfigProperty<String> INDEX_TYPE = ConfigProperty .key("hoodie.index.type") .noDefaultValue() .withDocumentation("Type of index to use. Default is Bloom filter. " + "Possible options are [BLOOM | GLOBAL_BLOOM |SIMPLE | GLOBAL_SIMPLE | INMEMORY | HBASE | BUCKET]. " + "Bloom filters removes the dependency on a external system " + "and is stored in the footer of the Parquet Data Files"); private String getDefaultIndexType(EngineType engineType) { switch (engineType) { case SPARK: return HoodieIndex.IndexType.SIMPLE.name(); case FLINK: case JAVA: return HoodieIndex.IndexType.INMEMORY.name(); default: throw new HoodieNotSupportedException("Unsupported engine " + engineType); } }` -- 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]
