Troubleshooting engines

If you are having trouble with an engine, here are some things that might help.
Check the logs
The default location for the engine logs is /var/log/heat/.
Restart the engine
In version 6.2.1.1 and later, to restart the engine and the related Heat API service, run the following commands:
systemctl restart openstack-heat-engine.service
systemctl restart openstack-heat-api.service
In versions before 6.2.1.1, to restart the engine and the related Heat API service, run the following commands:
service openstack-heat-engine restart
service openstack-heat-api restart
Verify the resource types
Verify that the engine has the resource types that are used by the blueprint design server by running the following command:
heat resource-type-list
If you see resource types that start with IBM::EC2 and IBM::UrbanCode, such as IBM::EC2::Port and IBM::UrbanCode::SoftwareDeploy::UCD, the engine has the correct resource types. If not, see Extending Heat orchestration engines.
Verify the allowed URLs for connections
Verify that the engine is allowed to access the Keystone server. Ensure that ports are open and network connectivity is available. Also, verify that the Keystone server is listed in the allowed_auth_uris parameter of the engine configuration file; see Connecting engines to Keystone servers.
Verify admin access to the heat stack domain
If you extended a Kilo heat engine and do not have admin permissions for your OpenStack-based cloud, you must obtain admin access to the heat domain of your heat engine before you provision an autoscaling group or deploy outside of IBM® UrbanCode™ Deploy. If you do not have admin access to the heat domain, when you provision an autoscaling group, the following error message is displayed:
KS-D4A211B You are not authorized to perform the requested action: identity:list_roles (HTTP 403)
From your engine administrator, obtain the following parameters and values for your OpenStack-based cloud:
stack_user_domain_name=heat
stack_domain_admin=heat_stack_admin
stack_domain_admin_password=heat_stack_admin_password
Locate each parameter in the engine heat.conf file, which is in the /etc/heat directory by default, and, if no value is provided, then enter the value that your engine administrator provided. You might need to uncomment the line of code that contains the parameter. If parameters are not present in the heat.conf file, then add the lines of code that contain them.

If the engine administrator cannot provide information for that domain, the engine administrator must create the domain. For example, to create the domain for a Kilo version engine on a Red Hat Enterprise Linux version 7, Fedora version 21, or CentOS version 7 operating system, see Install and configure Orchestration. Note that the process is different for each version of OpenStack. For example, for the equivalent process for a Mitaka version engine, see Install and configure

To connect to multiple OpenStack-based clouds by using admin access to the heat domain of an extended heat engine, you must use a separate engine for each OpenStack-based cloud.


Feedback