kfaraz commented on issue #17594: URL: https://github.com/apache/druid/issues/17594#issuecomment-2656301644
> Why did you remove the decommissioningMaxPercentOfMaxSegmentsToMove configuration that allowed us to configure the historical as deccommision, then set this config to 0 to tell the coordinator not to handoff all those segments. @BagamSingular , could you please elaborate how you were using this config? The `decommissioningNodes` field was intended to mark certain historicals as decommissioning so that all segments could be quickly moved away from them, as is currently done by Druid. You should not be using this field for any other purpose. Moreoever, the Coordinator only tries to rebalance segments if a server has gone offline. The `decommissioningMaxPercentOfMaxSegmentsToMove` played no part in it, as that applies only to live servers. In your case, IIUC, you restart a historical and it takes 20 mins to bootstrap. While the bootstrap is happening, the coordinator reassigns all the missing segments to other historicals. If this is indeed the case, then you may try using these configs for the time being: ``` smartSegmentLoading: false maxSegmentsInNodeLoadingQueue: 0 replicationThrottleLimit: 100 (or any other value that suits you depending on the total number of segments in your cluster) ``` -- 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]
