ActiveMQ uses an abstraction called a Discovery Agent to detect remote services such as remote brokers. We can use discovery for JMS clients to auto-detect a Message Broker to connect to, or to provide Networks of Brokers
There are currently two kinds of discovery agent.
The Discovery transport uses our own Multicast based discovery agent to locate the list of URIs to connect to.
For more information see the Discovery Transport Reference.
Zeroconf is a standard discovery specification that uses UDP / multicast to discovery devices. Its used by Apple's Rendezvous services.
If you have one or more brokers running with Zeroconf discovery enabled you can connect to a broker using the brokerURL
This will use Zeroconf to find an available broker and one will be randomly chosen & things will auto-failover on disconnect if there are several brokers running.
ActiveMQ supports the use of LDAP for discovery of brokers.
Please see LDAP Broker Discovery Mechanism for more details.
Trying out discovery
If you run the following commands in separate shells you'll have 2 brokers auto-discovering themselves and 2 clients using fixed-URLs
If you want the clients to use discovery to find brokers, run either of the two 'server' statements above (or both) then run the producer/consumer as follows
The transport URL is of the format
where <serviceName> is the Zeroconf service name; which seems to start with an underscore (_) and must end with a dot (.). So we can use this service name to distinguish development, UAT & production brokers - or group them into domains etc.
Discovery and Security
When using auto discovery of brokers an attacker may be able to present itself as a legitimate broker and by this way catch and / or manipulate all messages that run over it.
Are there security settings in auto discovery to avoid this?