Request among nodes taking minimum of idleConnectionTimeout configuration in tcpcommunicationSpi

[trans]

Hi Guys,

We have 4 clients(p1,p2,p3,p4) and 1 server running in Prague, then one
other node(i1) running in India.

But when we communicate from node in India to node in Prague then we see
that intermittently the request reaches to Prague after delay of around
30sec.
Node i1 logs has information that request was once tried and then next one
was tried after exact 30 sec.
idleConnectionTimeout is configured to 30s, so we think it may be related to
that.

Here is the configuration:

<beans xmlns="http://www.springframework.org/schema/beans";
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; xsi:schemaLocation="      
 
http://www.springframework.org/schema/beans       
http://www.springframework.org/schema/beans/spring-beans.xsd";>
  <bean class="org.apache.ignite.spi.discovery.tcp.TcpDiscoverySpi">
    <property name="joinTimeout" value="15000" />
    <property name="socketTimeout" value="14000" />
    <property name="ClientReconnectDisabled" value="true" />
    
    <property name="localPort" value="48500" />
    
    <property name="localPortRange" value="20" />
    <property name="ipFinder">
      <bean
class="org.apache.ignite.spi.discovery.tcp.ipfinder.vm.TcpDiscoveryVmIpFinder">
        <property name="addresses">
          <list>
            <value>myPragueMachine4660:48500..48520</value>
       
          </list>
        </property>
      </bean>
    </property>
  </bean>
  
  <bean id="communicationSpiBean"
class="org.apache.ignite.spi.communication.tcp.TcpCommunicationSpi">
    
    
    <property name="localPort" value="48100" />
    <property name="slowClientQueueLimit" value="100000" />
    <property name="idleConnectionTimeout" value="30000" />
    <property name="usePairedConnections" value="true" />
  </bean>
</beans>

Then after adding India machine entry in discovery ip, things started
working fine but we started seeing nearly twice of nodes in of topology, as
2 server and 8+ clients.

 <property name="ipFinder">
      <bean
class="org.apache.ignite.spi.discovery.tcp.ipfinder.vm.TcpDiscoveryVmIpFinder">
        <property name="addresses">
          <list>
            <value>myPragueMachine:48500..48520</value>
            <value>myIndiaMachine:48500..48520</value>
  
          </list>
        </property>
      </bean>
    </property>

Can someone please clarify
1. why our request was failing intermittently and then after exactly 30 sec
of time it reaches to Prague machine again. It looks like first requests'
TCP communication was idle and timedout then another request went. If yes,
then how can we be sure about the reason.
2. Why the topology has multiple nodes once we add the India Machine info in
TCPDiscoverySPI.
3. Should we add IPs of machine in both Prague and India in TCPDiscoverySPI.

Thanks.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/