[jira] [Created] (IGNITE-9135) TcpDiscovery - High Workload in Stable topology

[Anton Vinogradov (Jira)]

Sergey Kosarev created IGNITE-9135:
--------------------------------------

             Summary: TcpDiscovery - High Workload in Stable topology
                 Key: IGNITE-9135
                 URL: https://issues.apache.org/jira/browse/IGNITE-9135
             Project: Ignite
          Issue Type: Bug
            Reporter: Sergey Kosarev


On High topology (about 200 servers/ 50 clients) we see often  via jmx (TcpDiscoverySpiMBean) high MessageWorkerQueueSize peaks (>100) in stable cluster topology. Also very high number (about 250000) of ProcesedMessages, ReceivedMessages for TcpDiscoveryStatusCheckMessage, whereas TcpDiscoveryMetricsUpdateMessage is about 110000.

it looks like

org.apache.ignite.spi.discovery.tcp.ServerImpl.RingMessageWorker#metricsCheckFreq

value does not depend on topology size

private long metricsCheckFreq = 3 * spi.metricsUpdateFreq + 50;

 

 

 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)