nodejs-ClusterClientNoResponseError-client-no-response.md

• Node Version: 6.9.4 npm 4.1.1
• Egg Version: 0.12.0
• Plugin Name:
• Plugin Version:
• Platform: centos

egg.js部署在alpha环境下，发现时常出现卡的现象，首次卡差不多在5-20秒之间。查看日志记录如下：

• egg-web.log

2017-02-27 10:40:19,002 INFO 25022 [tcp-base] the connection: 127.0.0.1:43504 is closed by other side
2017-02-27 10:40:19,004 WARN 25022 [ClusterClient#Watcher] follower closed, and try to init it again
2017-02-27 10:40:19,017 INFO 25022 [ClusterClient#Watcher] failed to seize port 43504, and this is follower client.
2017-02-27 10:40:19,020 INFO 25022 [Follower#Watcher] register to channel: Watcher success
2017-02-27 10:41:40,605 INFO 25022 [tcp-base] the connection: 127.0.0.1:43504 is closed by other side
2017-02-27 10:41:40,606 WARN 25022 [ClusterClient#Watcher] follower closed, and try to init it again
2017-02-27 10:41:40,611 INFO 25022 [ClusterClient#Watcher] failed to seize port 43504, and this is follower client.
2017-02-27 10:41:40,613 INFO 25022 [Follower#Watcher] register to channel: Watcher success
2017-02-27 10:43:05,202 INFO 25022 [tcp-base] the connection: 127.0.0.1:43504 is closed by other side
2017-02-27 10:43:05,204 WARN 25022 [ClusterClient#Watcher] follower closed, and try to init it again
2017-02-27 10:43:05,216 INFO 25022 [ClusterClient#Watcher] failed to seize port 43504, and this is follower client.
2017-02-27 10:43:05,219 INFO 25022 [Follower#Watcher] register to channel: Watcher success
2017-02-27 10:44:36,837 INFO 25022 [tcp-base] the connection: 127.0.0.1:43504 is closed by other side
2017-02-27 10:44:36,838 WARN 25022 [ClusterClient#Watcher] follower closed, and try to init it again
2017-02-27 10:44:36,845 INFO 25022 [ClusterClient#Watcher] failed to seize port 43504, and this is follower client.
2017-02-27 10:44:36,847 INFO 25022 [Follower#Watcher] register to channel: Watcher success

egg-agent.log

2017-02-27 10:27:44,949 INFO 25012 [Leader#Watcher] socket connected, port: 52294
2017-02-27 10:29:12,108 INFO 25012 [Leader#Watcher] socket connected, port: 52332
2017-02-27 10:30:35,432 INFO 25012 [Leader#Watcher] socket connected, port: 52398
2017-02-27 10:32:01,630 INFO 25012 [Leader#Watcher] socket connected, port: 52584
2017-02-27 10:33:23,592 INFO 25012 [Leader#Watcher] socket connected, port: 52618
2017-02-27 10:34:47,405 INFO 25012 [Leader#Watcher] socket connected, port: 52628
2017-02-27 10:36:10,506 INFO 25012 [Leader#Watcher] socket connected, port: 52680
2017-02-27 10:37:32,651 INFO 25012 [Leader#Watcher] socket connected, port: 52690
2017-02-27 10:38:55,057 INFO 25012 [Leader#Watcher] socket connected, port: 52744
2017-02-27 10:40:19,019 INFO 25012 [Leader#Watcher] socket connected, port: 52784
2017-02-27 10:41:40,612 INFO 25012 [Leader#Watcher] socket connected, port: 53028
2017-02-27 10:43:05,218 INFO 25012 [Leader#Watcher] socket connected, port: 53096
2017-02-27 10:44:36,846 INFO 25012 [Leader#Watcher] socket connected, port: 53196

common-error.log

2017-02-27 00:01:25,445 ERROR 25012 nodejs.ClusterClientNoResponseError: client no response in 12397ms, maybe the connection is close on other side.
    at Timeout.Leader._heartbeatTimer.setInterval [as _onTimeout] (/pathxx/node_modules/cluster-client/lib/leader.js:59:23)
    at ontimeout (timers.js:365:14)
    at tryOnTimeout (timers.js:237:5)
    at Timer.listOnTimeout (timers.js:207:5)
name: 'ClusterClientNoResponseError'
pid: 25012
hostname: host-192-168-51-80

2017-02-27 00:02:45,501 ERROR 25012 nodejs.ClusterClientNoResponseError: client no response in 17120ms, maybe the connection is close on other side.
    at Timeout.Leader._heartbeatTimer.setInterval [as _onTimeout] (/pathxx/node_modules/cluster-client/lib/leader.js:59:23)
    at ontimeout (timers.js:365:14)
    at tryOnTimeout (timers.js:237:5)
    at Timer.listOnTimeout (timers.js:207:5)
name: 'ClusterClientNoResponseError'
pid: 25012
hostname: host-192-168-51-80

此日志文件记录内容最多

web.log 此日志文件为空

启动命令为

nohup node index.js > np-show20170220.log 2>&1 &

[kylezhang]

依然还有这个问题。

[itsky365]

@atian25

[root@host-192-168-1-1 code]# cat common-error.log 
2017-03-14 13:28:13,655 ERROR 28134 nodejs.ClusterClientNoResponseError: client no response in 31752ms exceeding maxIdleTime 30000ms, maybe the connection is close on other side.
    at Timeout.Leader._heartbeatTimer.setInterval [as _onTimeout] (/root/code/node_modules/.1.5.1@cluster-client/lib/leader.js:63:23)
    at ontimeout (timers.js:365:14)
    at tryOnTimeout (timers.js:237:5)
    at Timer.listOnTimeout (timers.js:207:5)
name: 'ClusterClientNoResponseError'
pid: 28134
hostname: host-192-168-1-1

现在还报这样的错误，只是比 egg@rc-2 、cluster-client@1.5.0 少些

运行环境 egg@rc-3 cluster-client@1.5.1