Cczzzz opened a new issue #3350: URL: https://github.com/apache/rocketmq/issues/3350
在服务端总有莫名其妙的超时情况发生: 2021-09-13 01:44:35 WARN SendMessageThread_14 - putMessage not in lock elapsed time(ms)=980, bodyLength=22 2021-09-13 01:44:35 WARN SendMessageThread_1 - putMessage not in lock elapsed time(ms)=981, bodyLength=6864 2021-09-13 01:44:35 WARN SendMessageThread_9 - putMessage not in lock elapsed time(ms)=982, bodyLength=6786 但是并没有[NOTIFYME]putMessage in lock cost time(ms) ，说明没有写入page cache 耗时很高的情况。可能是主从同步重新问题？ 场景： > tps <1k > 4.7.0 on dledger >16c 32g > 大概3天出现一次 今天发现 出现异常时有磁盘io读取和page in 的峰值，还有出现 meory page faults 的峰值。 这因为拉取历史消息发生缺页然后需要加载整个page cache 到内存中吗，这个行为会堵塞其他的写或者读吗。 日志中可以看见 Offset not matched. Request offset: 447420757556, firstOffset: 841813590016, lastOffset: 920196743168, mappedFileSize: 1073741824, mappedFiles count: 73 说明是一直有人缺拉取历史数据的。 读取历史数据会很严重的影响写入吗 -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
