J-HowHuang opened a new issue, #16327: URL: https://github.com/apache/pinot/issues/16327
Currently the remaining time for broker reduce phase is calculated as: the total query timeout, by table config or query option, subtracted by the time spent in waiting for final response from servers https://github.com/apache/pinot/blob/497a90b725bd60f45cceb9522a31c2f85839d10b/pinot-broker/src/main/java/org/apache/pinot/broker/requesthandler/SingleConnectionBrokerRequestHandler.java#L142 and then passed to `DataTableReducerContext` https://github.com/apache/pinot/blob/bc567a02b884b115d15351aada46305bf6df414c/pinot-core/src/main/java/org/apache/pinot/core/query/reduce/BrokerReduceService.java#L160-L162 However, only `GroupByDataTableReducer` has ever referenced to this value, that is, we could see `AggregationDataTableReducer`, `DistinctDataTableReducer` executed beyond the timeout time. For example, in the following log, you see `timeMs=7617`, `brokerReduceTimeMs=2864`, while query being `select percentile(_delivery_fee,50) from events_REALTIME limit 100000 OPTION(timeoutMs=4750);` ``` 2025/07/10 18:10:56.487 INFO [QueryLogger] [jersey-server-managed-async-executor-1] [cid=1538534192000001388] requestId=1538534192000001388,table=events_REALTIME,timeMs=7617,docs=22157799/22157794,entries=0/22157799,segments(queried/processed/matched/consumingQueried/consumingProcessed/consumingMatched/unavailable):148/148/148/1/1/1/0,consumingFreshnessTimeMs=1752171017918,servers=1/3,groupLimitReached=false,groupWarningLimitReached=false,brokerReduceTimeMs=2864,exceptions=1,serverStats=(Server=SubmitDelayMs,ResponseDelayMs,ResponseSize,DeserializationTimeMs,RequestSentDelayMs);pinot-server-server-0-6_R=0,-1,0,0,1;pinot-server-server-0-3_R=0,4621,177262737,63,1;pinot-server-server-0-7_R=1,-1,0,0,0,offlineThreadCpuTimeNs(total/thread/sysActivity/resSer):0/0/0/0,realtimeThreadCpuTimeNs(total/thread/sysActivity/resSer):1190142322/1145639363/0/44502959,clientIp=unknown,queryEngine=singleStage,offlineMemAllocatedBytes(total/thread/resSer):0/0/0,realtimeMemAllocatedBytes(total/thread/re sSer):0/0/0,query=select percentile(_delivery_fee,50) from events_REALTIME limit 100000 OPTION(timeoutMs=4750); 2025/07/10 18:10:56.487 ERROR [PinotClientRequest] [jersey-server-managed-async-executor-1] Query processing exceptions: {427=2 servers [pinot-server-server-0-6_R, pinot-server-server-0-7_R] not responded} ``` Even in `GroupByDataTableReducer`, the timeout is only checked in `getIndexedTable` https://github.com/apache/pinot/blob/767e3cddd6472bb12307af3e78f82d471be1559d/pinot-core/src/main/java/org/apache/pinot/core/query/reduce/GroupByDataTableReducer.java#L230-L232 -- 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]
