You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hudi.apache.org by "nsivabalan (via GitHub)" <gi...@apache.org> on 2026/06/25 03:23:13 UTC

Re: [I] Defer Initialization of MDT just at the end of first commit [hudi]

nsivabalan commented on issue #16059:
URL: https://github.com/apache/hudi/issues/16059#issuecomment-4795526072

   This appears to have been addressed by PR #18353 (commit `69fa35b`), which introduced the opt-in config `hoodie.metadata.record.level.index.defer.init` (default: `false`, advanced, since 1.2.0). When enabled, RLI initialization is deferred to the second commit on fresh tables, allowing the file group count to be determined dynamically from actual data instead of using the default of 10.
   
   Relevant code in `HoodieBackedTableMetadataWriter.java`:
   ```java
   // For a fresh table, defer RLI initialization
   if (dataWriteConfig.getMetadataConfig().shouldDeferRliInitForFreshTable()
       && this.enabledPartitionTypes.contains(RECORD_INDEX)
       && dataMetaClient.getActiveTimeline().filterCompletedInstants().countInstants() == 0) {
     this.enabledPartitionTypes.remove(RECORD_INDEX);
     partitionsToInit.remove(RECORD_INDEX);
   }
   ```
   
   A follow-up (PR #18399) rejects this config for the Flink writer. Closing as resolved — please reopen if a different scenario is still affected.


-- 
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: commits-unsubscribe@hudi.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org