PoC that creates brokers and interconnect routers. The brokers are in a master-slave configuration using replication (no shared store)
Follow the instructions for installing Ansible located here
For RHEL / Fedora, ensure the EPEL repository is configured, and then run:
sudo yum install ansible
The current Ansible playbook uses 5 AWS instances. Create 5 RHEL 7.2 instances (AMI ID: ami-775e4f16). Give each of these an elastic IP to prevent the IP addresses from changing.
git clone <repo>
Using the public DNS of the 5 servers provisioned above, update the hosts file. There will be 3 routers and 2 brokers in the hosts file. Set the ansible_host variable for each of the servers.
Additionally, set the ansible_ssh_private_key_file to the AWS pem file. If each host has a different pem file, this can be set on a per-host basis as follows:
routerA ansible_host=<host> ansible_ssh_private_key_file=<key>
The following command will set up both the brokers and the routers (the -T flag adds a connection timeout):
ansible-playbook site.yml -i hosts -T 30
If it is desired to run just one or the other, run one of the following commands:
ansible-playbook router.yml -i hosts -T 30
ansible-playbook broker.yml -i hosts -T 30
Navigate to the broker's Hawtio console (user:admin | password: admin)
http://<broker_master>:8161/hawtio
and verify that the broker has started successfully
There are two python scripts in this directory, send.py and receive.py. Both of these scripts use the following environment variables as input (shown below are the variables and their defaults):
MESSAGING_SERVICE_HOST=127.0.0.1
MESSAGING_SERVICE_PORT=5672
MESSAGE_RATE=200
MESSAGE_ADDRESS=jms.queue.testqueue
these defaults are all acceptable to start with, except for the host, which should be set to one of the ec2 router hosts as follows:
export MESSAGING_SERVICE_HOST=<ec2 router host>
To run the sender and receiver, run the following commands in two terminal windows:
python send.py
python receive.py
Messages will be enqueued by the artemis broker, and then the receive script will dequeue the messages. Two log files will be created, one named sender.log and one named receiver.log. These contain the message sequence of each sent and received message, respectively. To verify that no messages were lost, stop the send.py script and wait for the messages to drain (see the hawtio console to view the queue depth) and then run the following command, which will sort and diff the send and receive logs:
diff <(sort receiver.log) <(sort sender.log)
At this point, you should be able to send and receive messages using the python scripts, and kill the master broker and verify that messages continue to flow without message loss. Then, bring the master back online and verify it resumes as master and no messages are lost.