capistrant commented on code in PR #18844: URL: https://github.com/apache/druid/pull/18844#discussion_r2691759645
########## docs/data-management/automatic-compaction.md: ########## @@ -241,12 +241,18 @@ You can run automatic compaction using compaction supervisors on the Overlord ra * Can use either the native compaction engine or the [MSQ task engine](#use-msq-for-auto-compaction) * More reactive and submits tasks as soon as a compaction slot is available * Tracked compaction task status to avoid re-compacting an interval repeatedly +* Uses new Indexing State Fingerprinting mechanisms to store less data per segment in metadata storage -To use compaction supervisors, update the [compaction dynamic config](../api-reference/automatic-compaction-api.md#update-cluster-level-compaction-config) and set: +To use compaction supervisors, the following configuration requirements must be met: -* `useSupervisors` to `true` so that compaction tasks can be run as supervisor tasks -* `engine` to `msq` to use the MSQ task engine as the compaction engine or to `native` (default value) to use the native engine. +* You mut be using incremental segment metadata caching: + * `druid.manager.segments.useIncrementalCache` set to `always` or `ifSynced` in your Overlord runtime properties. Review Comment: this is true for both coordinator and overlord runtime props since the coordinator marks compaction fingerprints as active as part of segment insertion. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
