0

Artemis is configured in HA, with one master and a backup slave in different VM's.

The broker.xml configuration uses static discovery options and the acceptors, 61616 supporing all the protocols, and AMQP.

The configuration is similar to this link

artemis.log (below issue is continuously logged)

WARN [org.apache.activemq.artemis.core.client] AMQ212037: Connection failure to <ip-address> has been detected: readAddress(..) failed: Connection reset by peer [code=GENERIC_EXCEPTION]

From the logs, the ip-address info it seems to be one of the clients/consumer is pushing this info.

In another post link it was recommended to use confirmationWindowSize. But not sure whether this is applicable for my case.

The client configuration is using Camel and the ActiveMQ OpenWire client version 5.13.2 to connect:

    <bean id="conxnFactory" class="org.apache.activemq.ActiveMQConnectionFactory">
      <property name="brokerURL" value="tcp://domain-name:port" />
    </bean>

    <bean id="pooledconxnFactory" class="org.apache.activemq.pool.PooledConnectionFactory" init-method="start" destroy-method="stop">
      <property name="maxConnections" value="5" />
      <property name="connectionFactory" ref="conxnFactory" />
    </bean>

    <bean id="config" class="org.apache.camel.component.jms.JmsConfiguration">
      <property name="connectionFactory" ref="pooledconxnFactory" />
      <property name="concurrentConsumers" value="5" />
    </bean>

    <bean id="jmsqueue" class="org.apache.activemq.camel.component.ActiveMQComponent">
      <property name="configuration" ref="config" />
    </bean>
    
    <camelContext id="camel" xmlns="http://camel.apache.org/schema/spring">
      <endpoint id="inputQueue" uri="jmsqueue:queue:dataqueue1" />
      
      <route>
        <from uri="ref:inputQueue" />
        <!-- process -->
        <to uri="ref:outputQueue"/> <!-- endpoint defined already not included in here-->
        ...
        ...

Any inputs/ideas, why this occurs.

Justin Bertram
  • 29,372
  • 4
  • 21
  • 43
Tim
  • 1,321
  • 1
  • 22
  • 47
  • If the broker is reporting this it seems like there may be an issue with the client. See [this answer](https://stackoverflow.com/questions/1434451/what-does-connection-reset-by-peer-mean) for more details on what the error actually means. Is there a specific reason you're using the OpenWire client? Also, what version of the client are you using? – Justin Bertram Aug 03 '20 at 19:26
  • The recommendation you cited about using `confirmationWindowSize` will almost certainly *not* help in this case. The recommendation was based on the use of the HTTP transport with the Artemis core JMS client. You are not using either of these. – Justin Bertram Aug 03 '20 at 19:30
  • In our case we don't want the change the client side configuration due to legacy code complication. So we retained the apache camel 2.20.0 version and client info is similar as mentioned. I think that uses the OpenWire protocol not AMQP. I need to check if it can be modified to use AMPQ without much changes. – Tim Aug 03 '20 at 20:02
  • openwire client is activemq-client 5.13.2 version. – Tim Aug 03 '20 at 20:09
  • Is this something you can reproduce? If so, could you work up a [minimal, reproducible example](https://stackoverflow.com/help/minimal-reproducible-example)? – Justin Bertram Aug 03 '20 at 20:18
  • Will check the link you shared, and provide steps or info to reproduce this. – Tim Aug 03 '20 at 20:59

0 Answers0