-
|
看了很久没看出来哪里有问题, 也定位不了究竟是哪个Runnable超时了. broker返回异常org.apache.rocketmq.client.exception.MQBrokerException: CODE: 2 DESC: [TIMEOUT_CLEAN_QUEUE]broker busy, start flow control for a while, period in queue: 1322ms, size of queue: 0 BROKER: 10.2.0.4:10911 栈信息 |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 6 replies
-
|
这里是是服务端 Broker 因为 busy 而快速失败返回给客户端的，具体方法你可以看下 cleanExpiredRequestInQueue，如果队头任务等待时间超过了配置值（默认为200ms）就会 fast fail，以便客户端快速到其他节点重试 一般常见于发送队列，对于发送，最可能的就是 操作系统 pagecache 进行同步回收，阻塞了写入，然后导致写入队列堆积，可以看下服务端 store.log 中有没有以下日志来判断 |
Beta Was this translation helpful? Give feedback.
-
|
没有打印这个日志 2024-05-22 13:19:00 INFO AdminBrokerThread_4 - findPutMessageEntireTimePX 0.99=0.00ms cost 0ms |
Beta Was this translation helpful? Give feedback.
-
|
那这个问题如何解决呢？ |
Beta Was this translation helpful? Give feedback.
-
|
没解决, 开源的rocketmq社区活跃太差, 换成rabbitmq了 |
Beta Was this translation helpful? Give feedback.
-
我猜：试试ssd |
Beta Was this translation helpful? Give feedback.
-
SSD也试了, 没用 |
Beta Was this translation helpful? Give feedback.
-
|
发下这两个信息。
|
Beta Was this translation helpful? Give feedback.
-
感谢回复, 因为有点久远, stats.log日志已经找不到, 只有启动参数了 25 org.apache.rocketmq.broker.BrokerStartup -Xms8002M -Xmx8002M -Xmn1200M -XX:+UseG1GC -XX:G1HeapRegionSize=16m -XX:G1ReservePercent=25 -XX:InitiatingHeapOccupancyPercent=30 -XX:SoftRefLRUPolicyMSPerMB=0 -XX:SurvivorRatio=8 -verbose:gc -Xloggc:/dev/shm/mq_gc_%p.log -XX:+PrintGCDetails -XX:+PrintGCDateStamps -XX:+PrintGCApplicationStoppedTime -XX:+PrintAdaptiveSizePolicy -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=5 -XX:GCLogFileSize=30m -XX:-OmitStackTraceInFastThrow -XX:+AlwaysPreTouch -XX:MaxDirectMemorySize=8002M -XX:-UseLargePages -XX:-UseBiasedLocking -Djava.ext.dirs=/opt/java/openjdk/jre/lib/ext:/home/rocketmq/rocketmq-5.3.0/bin/../lib -Xms2g -Xmx2g -XX:+AlwaysPreTouch -Drmq.logback.configurationFile=/home/rocketmq/rocketmq-5.3.0/conf/rmq.broker.logback.xml |
Beta Was this translation helpful? Give feedback.
这里是是服务端 Broker 因为 busy 而快速失败返回给客户端的，具体方法你可以看下 cleanExpiredRequestInQueue，如果队头任务等待时间超过了配置值（默认为200ms）就会 fast fail，以便客户端快速到其他节点重试
一般常见于发送队列，对于发送，最可能的就是 操作系统 pagecache 进行同步回收，阻塞了写入，然后导致写入队列堆积，可以看下服务端 store.log 中有没有以下日志来判断