openstack/openstack-ansible
Code Issues Proposed changes

Set Rabbit recovery method for network partitions

Browse Source 
Adds configurable option rabbit_cluster_partition_handling with default
value of pause_minority. Allows option to be explicitly set to ignore
through Ansible to reset to prior (and RabbitMQ default) behavior.

Change-Id: I52625390b388a55c6ed210b5804c7de336985f37
Closes-Bug: #1454860
This commit is contained in:
Steve Lewis 2015-05-27 18:39:48 -07:00
parent 21575644b1
commit af1b69cb1e
2 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
playbooks/roles/rabbitmq_server/defaults/main.yml
View File

@ -34,6 +34,9 @@ rabbitmq_userid: rabbitmq
# Name of the rabbitmq cluster
rabbitmq_cluster_name: rabbitmq_cluster1
# Specify a partition recovery strategy (autoheal | pause_minority | ignore)
rabbitmq_cluster_partition_handling: pause_minority
# Rabbitmq open file limits
rabbitmq_ulimit: 4096

1
playbooks/roles/rabbitmq_server/templates/rabbitmq.config.j2
View File

@ -1,6 +1,7 @@
[
{rabbit, [
{loopback_users, []},
{% if rabbitmq_cluster_partition_handling != 'ignore' %}{cluster_partition_handling, {{ rabbitmq_cluster_partition_handling }}},{% endif %}
{cluster_nodes, {
[ {% for host in groups['rabbitmq_all'] %}'rabbit@{{ hostvars[host]['ansible_ssh_host'] }}'{% if not loop.last %}, {% endif %}{% endfor %}], disc}
}