You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@rocketmq.apache.org by GitBox <gi...@apache.org> on 2020/04/10 08:00:09 UTC

[GitHub] [rocketmq] Kanepan commented on issue #348: [TIMEOUT_CLEAN_QUEUE]broker busy

Kanepan commented on issue #348: [TIMEOUT_CLEAN_QUEUE]broker busy
URL: https://github.com/apache/rocketmq/issues/348#issuecomment-611926018
 
 
   > > 在tps不到100的情况下,还出现serverbusy,你的服务器的配置可能不是很高,sendMessageThreadPoolNums这个值不宜配置这么多,另外lizhanhui大神已经建议过,在同步刷盘模式下,建议把waitTimeMillsInSendQueue适当调整大一点。
   > > 可以试试下面的参数
   > > sendMessageThreadPoolNums=16
   > > useReentrantLockWhenPutMessage=true
   > > waitTimeMillsInSendQueue=500
   > > 另外,如果想再提高TPS,建议增加下面几个配置
   > > transientStorePoolEnable=true
   > > transientStorePoolSize=2
   > > 如果你的服务器内存还比较富余,transientStorePoolSize可以配置到5
   > > 如果尝试配置还不能解决你的问题,烦请下次把你broker的所有配置都贴出来!
   > 
   > 你好,我们MQ 机器是 4C 32G 的虚拟机.
   > `vm.swappiness=1`
   > 
   > JVM 配置:
   > `-Xmx10G -Xms10G -XX:+UseG1GC `
   > (这里只列出主要参数)
   > 
   > broker配置:
   > 
   > ```
   > brokerRole=ASYNC_MASTER 
   > flushDiskType=ASYNC_FLUSH
   > sendMessageThreadPoolNums=32
   > useReentrantLockWhenPutMessage=true
   > waitTimeMillsInSendQueue=400
   > transientStorePoolEnable=true
   > ```
   > 
   > TPS 300的情况下,偶尔还会报 broker busy 和 system busy。
   > 报错期间看监控,CPU 和 IO 都不高,请问参数有什么问题吗? 哪里需要调整?非常感谢
   
   waitTimeMillsInSendQueue=6000
   osPageCacheBusyTimeOutMills=6000
   按这样设置, 普通的云服务器磁盘IO都比较弱, ROKCET官方出厂的设置比较敏感,如果解决问题了麻烦反馈下。

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services