ActiveMQ启动时出现tcp连接失败(java.io.EOFException)问题求助

错误日志 #

2022-07-18 14:02:48,968 | WARN | Transport Connection to: tcp://127.0.0.1:58090 failed: java.io.EOFException | org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ Transport: tcp:///127.0.0.1:58090@61616
2022-07-18 14:02:48,972 | WARN | Transport Connection to: tcp://127.0.0.1:58094 failed: java.io.EOFException | org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ Transport: tcp:///127.0.0.1:58094@61616
2022-07-18 14:02:48,976 | WARN | Transport Connection to: tcp://127.0.0.1:58098 failed: java.io.EOFException | org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ Transport: tcp:///127.0.0.1:58098@61616
2022-07-18 14:02:48,983 | WARN | Transport Connection to: tcp://127.0.0.1:58102 failed: java.io.EOFException | org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ Transport: tcp:///127.0.0.1:58102@61616
2022-07-18 14:02:48,994 | WARN | Transport Connection to: tcp://127.0.0.1:58106 failed: java.io.EOFException | org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ Transport: tcp:///127.0.0.1:58106@61616
2022-07-18 14:02:49,009 | WARN | Transport Connection to: tcp://127.0.0.1:58114 failed: java.io.EOFException | org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ Transport: tcp:///127.0.0.1:58114@61616
2022-07-18 14:02:49,009 | WARN | Transport Connection to: tcp://127.0.0.1:58110 failed: java.io.EOFException | org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ Transport: tcp:///127.0.0.1:58110@61616

已知信息 #

• 启动ActiveMQ时持续触发上述WARN日志，提示本地TCP连接因EOFException失败
• 最初排查磁盘空间占用超90%，扩容后问题未解决
• 运行环境：Ubuntu系统，搭配Atlassian旗下PIM解决方案Contentserv

排查与解决建议 #

• 调整ActiveMQ传输连接器的心跳配置
  编辑activemq.xml中的TCP连接器配置，添加空闲超时参数，避免无数据传输时被误判为连接失效：

  <transportConnector name="tcp" uri="tcp://0.0.0.0:61616?wireFormat.maxInactivityDuration=30000&amp;wireFormat.maxInactivityDurationInitalDelay=60000"/>
  

  重启ActiveMQ生效。

• 校验Contentserv客户端连接参数
  确认Contentserv侧配置的ActiveMQ连接参数（超时时间、重试策略等）与服务端匹配，避免客户端异常断开引发服务端EOF警告。

• 优化系统TCP连接回收

  1. 执行netstat -anp | grep 61616查看端口连接状态，确认是否存在大量TIME_WAIT/CLOSE_WAIT连接
  2. 编辑/etc/sysctl.conf添加以下参数：

     net.ipv4.tcp_tw_reuse = 1
     net.ipv4.tcp_fin_timeout = 30
     

  3. 执行sysctl -p使配置生效，加快TCP连接回收速度。

• 开启详细日志定位根因
  临时将ActiveMQ日志级别调整为DEBUG，查看连接建立、断开的完整流程，确认是客户端主动断开还是协议不兼容导致的EOF异常。

内容的提问来源于stack exchange，提问作者Hamza AZIZ