Can we use JIRA to track this kind of issue? It's easy to forget if we don't track it.
Willem Jiang Blog: http://willemjiang.blogspot.com (English) http://jnn.iteye.com (Chinese) Twitter: willemjiang Weibo: 姜宁willem On Wed, Mar 8, 2017 at 10:34 AM, Jie Tang <[email protected]> wrote: > @Rick Zhang > > That's a good suggestion! > The feature is useful. > We plan to add monitor(alert) in milestone 2. > (In fact,milestone 1 already has some code for this feature,but we don't > have enough time to develop it,see > org.apache.rocketmq.console.service.MonitorService) > > We can discuss what we want in RocketMQ-Console milestone 2. Waiting for > your suggestions. > > for example: > > monitor alert system (rocketMQ health check) > > auto ops(install/upgrade rockemtMQ broker/namesvr use console) > ... > > > And I hope more and more people can join us.if you are interested in > RocketMQ-Console,you can contribute something for it.We work together to > make it better. > > :-) > > Now milestone 1 of RocketMQ-Console is ready to release. > > enjoy it~ > > https://github.com/rocketmq/rocketmq-console-ng > > 2017-03-08 8:03 GMT+08:00 Rick Zhang <[email protected]>: > > > Hi RocketMQ Team, > > Now in RocketMQ, there is only a simple watermark, which print > runtime > > messages to a log file. In production, a more powerful monitor/alarm > > functionality is necessary. For instance, if messages are accumulated in > a > > queue, we should know the first time. Sometimes the accumulation is > normal, > > because MQ takes a responsibility of reservoir, but sometimes it means > > downstream consumes too slow or die that we have to check what happen. I > > know the alarm threshold is not easy to set, but even though a fixed > > threshold set manually according to experience is better than none. > > > > Alarm may should be a feature of RocketMQ-Console, but a powerful > > monitor may should be a part of RMQ. If you feel it's a useful feature, > we > > could discuss it further in dev@list. > > >
