We have an ActiveMQ instance set up with about 300 consumers reading from persistent queues. These consumers are outside our network, so we're using the HTTP tunneling feature to communicate with the queues, using the JMS API.
The problem we are seeing is that the consumers appear to be polling as fast as they get a response. The client we deploy starts six consumers each, and we want to be able to deploy more clients. Problem is, the more clients we deploy, the more traffic we see. Right now our servers are handling more than 600 HTTP requests per second. I started up an ActiveMQ MesageConsumer locally and found that it made about 50-60 requests per second, which seems...excessive. Is there a way to slow things down a bit? Looking through both the MQ and JMS documentation, I didn't see a way to change the poll interval, probably because it wasn't designed around HTTP tunneling. If not, is there a way to configure the MQ server so that it won't return as fast, also cutting down on the number of requests made? -- View this message in context: http://activemq.2283324.n4.nabble.com/Rate-limiting-client-polling-tp4666466.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
