Deployment fails if enable_outward_rabbitmq is set to 'yes'
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
kolla-ansible |
Fix Released
|
Medium
|
arlind |
Bug Description
Hi,
I have set enable_
outward_rabbitmq docker log contains this error:
init terminating in do_boot ({error,
I have a 2 node baremetal cluster:
- os-controller01 (controller, network, monitoring)
- os-cs01 (compute, storage)
I have attached a zip file with logs and configuration:
- doutward_rabbitmq container logs - https:/
- ansible output - https:/
- globals.yml - https:/
- multinode inventory file - https:/
- /var/log/
**Environment**:
* OS: CentOS Linux release 8.2.2004 (Core)
* Kernel: 4.18.0-
* Docker version: Docker Engine - Community - 19.03.11
* Kolla-Ansible version: stable/ussuri branch
* Docker image Install type: binary
* Docker image distribution: centos
Changed in kolla-ansible: | |
status: | New → Confirmed |
Changed in kolla-ansible: | |
importance: | Undecided → Medium |
assignee: | nobody → arlind (arlind12) |
Changed in kolla-ansible: | |
status: | Confirmed → In Progress |
Apparently this bug comes from enabling rabbitmq_prometheus plugin, which listens on port 15692 and is not customizable. outward_rabbitmq fails to start because port 15692 is already used by the first rabbitmq container. This value is not configurable and is hardcoded in the config file '/var/lib/ rabbitmq/ mnesia/ rabbit@ <node_name> -plugins- expand/ rabbitmq_ prometheus- 3.8.5/priv/ schema/ rabbitmq_ prometheus. schema'