Hello guys, I’ve optimized the queries — when I run them directly in the Pinot controller, they complete in about 100–150 ms, and through Java integration they usually take around 200 ms. However, sometimes the execution time spikes to over 10 seconds. Could you please help me understand what might be causing this — is it related to the query itself or the Pinot table configuration? I set timeout 10 sec. 2025-10-13 154731.738 log=[{"errorCode":200,"message":"QueryExecutionError:\nReceived error query execution result block: {250=ExecutionTimeoutError\nProcessingException(errorCode:250, message:ExecutionTimeoutError)\n\tat org.apache.pinot.common.exception.QueryException.<clinit>(QueryException.java:113)\n\tat org.apache.pinot.common.datablock.DataBlockUtils.extractErrorMsg(DataBlockUtils.java:55)\n\tat org.apache.pinot.common.datablock.DataBlockUtils.getErrorDataBlock(DataBlockUtils.java:47)\n\tat org.apache.pinot.query.runtime.blocks.TransferableBlockUtils.getErrorTransferableBlock(TransferableBlockUtils.java:54)}\norg.apache.pinot.query.service.dispatch.QueryDispatcher.runReducer(QueryDispatcher.java:306)\norg.apache.pinot.query.service.dispatch.QueryDispatcher.submitAndReduce(QueryDispatcher.java:96)\norg.apache.pinot.broker.requesthandler.MultiStageBrokerRequestHandler.handleRequest(MultiStageBrokerRequestHandler.java:219)\norg.apache.pinot.broker.requesthandler.BaseBrokerRequestHandler.handleRequest(BaseBrokerRequestHandler.java:133)\n"}] 2025-10-13 154731.739 log=101731 ERROR traceId=, parentId=, spanId=, sampled= [io.qu.mu.ru.MutinyInfrastructure] (executor-thread-33) Mutiny had to drop the following exception: io.fastpix.metrix.AppException: something went wrong in pinot 2025-10-13 154731.739 log= at io.quarkus.vertx.core.runtime.VertxCoreRecorder$15.runWith(VertxCoreRecorder.java:638) 2025-10-13 154731.739 log= at org.jboss.threads.EnhancedQueueExecutor$ThreadBody.run(EnhancedQueueExecutor.java:1594) 2025-10-13 154731.739 log= at org.jboss.threads.DelegatingRunnable.run(DelegatingRunnable.java:11) 2025-10-13 154731.739 log= at org.jboss.threads.ThreadLocalResettingRunnable.run(ThreadLocalResettingRunnable.java:11) 2025-10-13 154731.739 log= at org.jboss.threads.EnhancedQueueExecutor$Task.run(EnhancedQueueExecutor.java:2654) 2025-10-13 154731.739 log= at io.smallrye.context.impl.wrappers.SlowContextualSupplier.get(SlowContextualSupplier.java:21) 2025-10-13 154731.739 log= at java.base/java.lang.Thread.run(Thread.java:1583) 2025-10-13 154731.739 log= at io.smallrye.mutiny.operators.uni.UniRunSubscribeOn.lambda$subscribe$0(UniRunSubscribeOn.java:27) 2025-10-13 154731.739 log= at io.fastpix.metrix.utils.PinotClientConfig_ClientProxy.executeQueryAsync(Unknown Source) 2025-10-13 154731.739 log= at io.fastpix.metrix.services.impl.MetricServiceImpl.lambda$getMetricsOfBreakdown$1(MetricServiceImpl.java:420) 2025-10-13 154731.739 log= at org.jboss.threads.EnhancedQueueExecutor.runThreadBody(EnhancedQueueExecutor.java:1627) 2025-10-13 154731.739 log= at org.jboss.threads.EnhancedQueueExecutor$Task.doRunWith(EnhancedQueueExecutor.java:2675) 2025-10-13 154731.739 log= 2025-10-13 154731.739 log= at io.smallrye.mutiny.operators.AbstractUni.subscribe(AbstractUni.java:35) 2025-10-13 154731.739 log= at io.fastpix.metrix.utils.PinotClientConfig.executeQueryAsync(PinotClientConfig.java:61) 2025-10-13 154731.739 log= at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)

Is your java app in the same network / vpc, or querying via VPN, over network / region? Latter would add latency. If same query takes longer at times, the first thing to look at is whether there are other queries running, second thing is if the query is loading cold pages into memory for the first time (cold start)

Hello @Mayank, It’s running within the same network. It’s not just a single query — most of the optimized queries occasionally show the same behavior. The issue isn’t frequent, but it happens from time to time.

So no qps?

what is qps.

I meant are there multiple queries running when this happens

In our dashboard, we’re calling 5–8 APIs simultaneously, resulting in 5–8 parallel queries. The issue isn’t from the Java side — the timeout error is coming from Pinot.

Will need to check server / broker logs to figure out exactly what happened. You can start with broker logs for the query