Hotrod remote server exception

[jacob1111]

Hi,

I am trying to use Infinispan remote cache on the remote server. I am using Java application

to connect to the server and store objects in the memory.

When I ran this application for the local hotrod server using loopback address

(127.0.0.1) it works. However, when I try to use it for remote server it fails.

Here is the code snippet:

import org.infinispan.client.hotrod.RemoteCache;

import org.infinispan.client.hotrod.RemoteCacheManager;

public class HotRodRemoteClient

{

  public void start() 

  {

    RemoteCacheManager manager = new RemoteCacheManager("10.100.9.28");

    RemoteCache<Integer, Ticket> cache = manager.getCache();

     ............................................

Here is the exception:

ISPN004017: Could not fetch transport

org.infinispan.client.hotrod.exceptions.TransportException:: Could not connect to server: /10.100.9.28:11222

    at org.infinispan.client.hotrod.impl.transport.tcp.TcpTransport.<init>(TcpTransport.java:90)

    at org.infinispan.client.hotrod.impl.transport.tcp.TransportObjectFactory.makeObject(TransportObjectFactory.java:57)

    at org.apache.commons.pool.impl.GenericKeyedObjectPool.borrowObject(GenericKeyedObjectPool.java:1220)

    at org.infinispan.client.hotrod.impl.transport.tcp.TcpTransportFactory.borrowTransportFromPool(TcpTransportFactory.java:254)

    at org.infinispan.client.hotrod.impl.transport.tcp.TcpTransportFactory.getTransport(TcpTransportFactory.java:145)

    at org.infinispan.client.hotrod.impl.operations.FaultTolerantPingOperation.getTransport(FaultTolerantPingOperation.java:44)

    at org.infinispan.client.hotrod.impl.operations.RetryOnFailureOperation.execute(RetryOnFailureOperation.java:67)

    at org.infinispan.client.hotrod.impl.RemoteCacheImpl.ping(RemoteCacheImpl.java:432)

    at org.infinispan.client.hotrod.RemoteCacheManager.ping(RemoteCacheManager.java:538)

    at org.infinispan.client.hotrod.RemoteCacheManager.createRemoteCache(RemoteCacheManager.java:520)

    at org.infinispan.client.hotrod.RemoteCacheManager.getCache(RemoteCacheManager.java:452)

    at org.infinispan.client.hotrod.RemoteCacheManager.getCache(RemoteCacheManager.java:447)

    at com.packtpub.infinispan.chapter2.HotRodRemoteClient.start(HotRodRemoteClient.java:17)

    at com.packtpub.infinispan.chapter2.HotRodRemoteClient.main(HotRodRemoteClient.java:65)

Caused by: java.net.ConnectException: Connection refused

    at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method)

    at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:599)

    at sun.nio.ch.SocketAdaptor.connect(SocketAdaptor.java:100)

    at org.infinispan.client.hotrod.impl.transport.tcp.TcpTransport.<init>(TcpTransport.java:81)

    ... 13 more

I can ping the server 10.100.9.28:

shell> ping 10.100.9.28

PING 10.100.9.28 (10.100.9.28): 56 data bytes

64 bytes from 10.100.9.28: icmp_seq=0 ttl=64 time=0.261 ms

64 bytes from 10.100.9.28: icmp_seq=1 ttl=64 time=0.184 ms

64 bytes from 10.100.9.28: icmp_seq=2 ttl=64 time=0.290 ms

64 bytes from 10.100.9.28: icmp_seq=3 ttl=64 time=0.285 ms

I use Infinispan 5.1.6, Maven 3.04, JDK6.33.

My hotrod server runs on Centos 5.5, but Java application is on Mac 10.7

Firewalls on both machines are disabled.

I suspect this is configuration problem. What should I change to make it to work?

Thank you,

Jacob Nikom

[mircea.markus]

is the server started on 11222? can you telnet there?

[jacob1111]

Hi Mircea,

Thank you very much for your response.

I have serious difficulties trying to get data using HotRod server

that runs on the remote node - I mean on the machine that is

different than the HotRod client runs.

This is very imprtant case for me because I am trying to emulate

near cache use case.

The client program is very simple. When I run it on the local node

(the same as HotRod server) it works flawlessly.

If I ran HotRod server in on remote node the client cannot get the server.

In both cases HotRod servers on local and remote nodes work very well.

I don't have any problems with them.

I use Infinispan default settings, so it is likey I use jgroup_udp.xml

I did not change that file, but I had to change a little bit HotRod startup

script to get it to work.

Now it looks like that:

===================================================

shell> more startServer.sh.next

#!/bin/bash

source "`dirname "$0"`/functions.sh"

CLIENT_HOME=/home/jnikom/Kiva/Downloads/Java/Infinispan/dev/HotRod_01

add_classpath "${ISPN_HOME}"/*.jar

add_classpath "${ISPN_HOME}/lib"

add_classpath "${ISPN_HOME}/modules/memcached"

add_classpath "${ISPN_HOME}/modules/hotrod"

add_classpath "${ISPN_HOME}/modules/websocket"

add_jvm_args $JVM_PARAMS

add_jvm_args '-Djava.net.preferIPv4Stack=true'

add_jvm_args '-Djgroups.bind_addr=10.100.9.28'

add_jvm_args '-Djgroups.udp.mcast_addr=228.10.10.10'

add_jvm_args '-Djgroups.udp.mcast_port=45588'

add_jvm_args '-Djgroups.udp.ip_ttl=5'

add_jvm_args '-Xms512m'

add_jvm_args '-Xmx3g'

# RHQ monitoring options

add_jvm_args '-Dcom.sun.management.jmxremote.ssl=false'

add_jvm_args '-Dcom.sun.management.jmxremote.authenticate=false'

add_jvm_args -Dcom.sun.management.jmxremote.port=$(find_tcp_port)

# Workaround for JDK6 NPE: http://bugs.sun.com/view_bug.do?bug_id=6427854

add_jvm_args '-Dsun.nio.ch.bugLevel=""'

# Sample JPDA settings for remote socket debugging

#add_jvm_args "-Xrunjdwp:transport=dt_socket,address=8686,server=y,suspend=n"

# LOG4J configuration

LOG4J_CONFIG=file:///${CLIENT_HOME}/log4j.properties

add_program_args "$@"

start org.infinispan.server.core.Main

===========================================

Here is the snippet of the Java code:

===========================================

import java.util.Set;

import java.util.Map;

import java.util.Iterator;

import org.infinispan.client.hotrod.RemoteCache;

import org.infinispan.client.hotrod.RemoteCacheManager;

public class HotRodRemoteClient

{

  public void start() 

  {

    RemoteCacheManager rm = new RemoteCacheManager("127.0.0.1");

//  RemoteCacheManager rm = new RemoteCacheManager("10.100.9.9");

    RemoteCache<Integer, Ticket> rc = rm.getCache("clusteredCache");

    String command = ""; // Line read from standard in

    int ticketid = 1;

....................................

...................

====================================================

Again, HotRod servers on both nodes run very well.

I use Intel Zeon 2.0 GHz, 2 CPU 8 core hardware,

CentOS 5.5, JDK 1.6.0_31

Infinispan 5.1.6

My another node is a virtual system. I run CentOS 5.5

on Mac i7 using JDK 1.6.0_31 and Infinispan 5.1.6.

I am able to ssh over there, ping, telnet, etc.

Moreover, in embeeded mode I am able to replicate (Infinispan)

between those two machines.

If I am able to do that,  why I cannot reach the remote HorRod cerver

from my client?

Thanks again for your help.

Best regards,

Jacob Nikom

[jacob1111]

Hi Mircea,

I also wanted to add Infinispan configuration file for this problem:

==============================================================

<infinispan>

    <global>

        <transport nodeName="Infinispan-Node1" clusterName="infinispan-cluster" >

        </transport>

    </global>

    <namedCache name="clusteredCache">

        <clustering mode="replication">

             <stateTransfer

                chunkSize="0"

                fetchInMemoryState="false"

                timeout="240000">

             </stateTransfer>

             <sync replTimeout="20000"/>

          </clustering>

    </namedCache>

</infinispan>

===============================================

[mircea.markus]

I'm sorry I cannot be of more help Jacob, but this looks like a network releated issue to me. What HRClient does is simply:

         socketChannel = SocketChannel.open();

         socket = socketChannel.socket();

         socket.connect(serverAddress, transportFactory.getConnectTimeout());

Do you happen to hav multiple network interfaces? Might be that the client connects to the wrong one..

[galder.zamarreno]

Cross posted: http://stackoverflow.com/questions/12415537/infinispan-hotrod-remote-server-exception

[jacob1111]

Hi Galder,

Thank you for your reply.

Unfortunately, I could not find your comments about this case.

Did you answer my question?

Regards,

Jacob Nikom