So producer flow control works as follows: The producer has a window that gets smaller each time he sends a message. The client will stop sending once that window closes. When the broker has capacity, hes sends commands back to the producer to expand his window.
Regards, Hiram On Thu, Jun 5, 2008 at 6:04 AM, Madfrog <[EMAIL PROTECTED]> wrote: > > Dear sir: > Non-persistent messages will eventually fill up the memory made > available for queues, and persistent messages will run out of allocated > storage space and both situations pose risks to the stability of the message > server and the entire distributed system. In order to ensure stability, many > JMS vendors implement 'flow control' mechanisms that limit resource > usage under heavy load. > So, I search in the forum of ActiveMQ,and found this: > As of ActiveMQ 5.0, we can now individually flow control each > producer on a shared connection without having to suspend the entire > connection. Messages that are sent synchronously will automatically use per > producer flow control in > http://activemq.apache.org/producer-flow-control.html. > And I alse see the the ActiveMQMessageProducer code. > However, I can not understand the flow control mechanism in > ActiveMQ, > Could someone kind enough to help me? :) > -- > View this message in context: > http://www.nabble.com/How-Producer-Flow-Control-works-in-activeMQ-5.1-tp17666469s2354p17666469.html > Sent from the ActiveMQ - Dev mailing list archive at Nabble.com. > > -- Regards, Hiram Blog: http://hiramchirino.com Open Source SOA http://open.iona.com
