Connecting the blueprint design server to Amazon Web Services

To connect to Amazon Web Services (AWS), map the AWS account information to a functional ID. Then, assign that functional ID to a team.

Before you begin

Obtain an OpenStack Keystone server. The blueprint design server requires a Keystone server to connect to any cloud. You can install a Keystone server when you install the engine, reuse a Keystone server that is connected to a different cloud, or install a different Keystone server for use with AWS. See the OpenStack documentation.
Obtain an engine. The engine version must match the version of the OpenStack Keystone server. You can use any of the following options for the engine:
- Install an engine. See Installing engines in silent mode or Installing engines in interactive mode.
- If you already have an engine, connect it to the Keystone server as described in Connecting engines to Keystone servers.
- Extend an existing Heat orchestration engine and connect it to the Keystone server. See Extending Heat orchestration engines.
Create a functional user account on the Keystone server. This user account must be a member of the administrative tenant on the Keystone server. Later, you associate the AWS account information with this functional ID. With this account, users can authenticate to AWS.
Install the blueprint design server. See Installing the blueprint design server.
Connect the blueprint design server to the server. See Connecting the blueprint design server to the server.
Configure an authentication realm for the blueprint design server. You can import users from a variety of sources, including LDAP servers, Keystone identity services, the IBM® UrbanCode™ Deploy server, or from the internal authentication realm. See Creating authentication realms for the blueprint designer
Ensure that the blueprint design server can connect to AWS. You can verify the connection path with the curl or telnet commands. For example, make sure that no firewall, proxy, or security settings prevent communication between the blueprint design server and the cloud. The blueprint design server must be able to access the AWS endpoints for the regions that you plan to use, such as ec2.ap-northeast-1.amazonaws.com and ec2.us-west-2.amazonaws.com.

About this task

The following diagram shows a typical topology for this scenario. The blueprint design server and engine connect to Amazon Web Services. For authentication information, the blueprint design server connects to the Keystone identity service and optionally to an LDAP server.

A topology that includes the blueprint design server, an engine, Amazon Web Services, a Keystone server, and an optional LDAP server

Procedure

Log in to the blueprint designer as a user with the following System permissions:
- Configure Security
- Manage Users & Groups
Create a connection to the cloud:
1. Click Settings > Clouds.
2. Click Add New Cloud.
3. Specify a name for the cloud connection.
4. In the Type list, select Amazon Web Services.
5. In the Endpoint Type list, select the type of URL that you use to connect to this cloud.
  - If you connect through a private URL, select Internal.
  - If you connect through a public URL, select Public.
6. In the Identity URL field, specify the location of the identity service, such as http://example.com:5000/v2.0 or http://example.com:5000/v3. Do not include a trailing slash. If you installed a Keystone server along with the engine that is included with IBM UrbanCode Deploy, you can use that server. If you installed version 6.2.1.1 and later, specify the value http://engineHostname:5000/v3. If you installed a version before 6.2.1.1, specify the value http://engineHostname:5000/v2.0. In both examples, the value for engineHostname is the host name or IP address of the engine.
7. In the Timeout in Mins field, specify the amount of time in minutes to wait for a provision request to be completed. If you deploy IBM UrbanCode Deploy components, allow sufficient time for the cloud to provision your instance, the agent to come online, and all processes to run. See Creating a IBM UrbanCode Deploy timeout configuration file.
8. Specify the Heat orchestration engine to use:
  - If your Keystone server was not installed with an engine that was included with IBM UrbanCode Deploy, to use the default Heat engine for the Keystone server, select the Use default orchestration engine check box.
    Note: This engine must have the custom types for the blueprint design server as described in Extending Heat orchestration engines.
  - If your Keystone server was installed with an engine that was included with IBM UrbanCode Deploy, or to use a different Heat engine, clear the Use default orchestration engine check box and specify the location of your engine, such as http://engine.example.com:8004.
    Note: Do not use the localhost variable in this field, even if the engine is on the same system as the blueprint design server.
9. Optional: Select the cost center to use to estimate the cost of environments on this cloud.
10. Click Save.
Create one or more cloud projects that tie the functional ID on the Keystone server to the AWS account information. See Creating cloud projects for the blueprint designer.
Add the cloud project to a team.
Add users to the team and to one or more roles on the team. These users can come from any authentication realm, including LDAP servers, Keystone identity services, or from the internal authentication realm.
Make sure that the team roles include the appropriate permissions for those users, such as creating and editing blueprints.
Register the Amazon Elastic Compute Cloud (EC2) images with the cloud discovery service. See Registering Amazon EC2 images with the cloud discovery service.

Results

Users can log in to the blueprint designer and use the cloud connection. At the top of the page, users can select the AWS cloud connection, cloud project, and region. When they edit blueprints, the palette shows resources that are available to the AWS account, and they can provision blueprints to the selected region.

Feedback