Setting Up VMRs in OpenStack

This tutorial will walk you through the steps required to get a single Solace Virtual Message Router (VMR) instance running in OpenStack and ready for messaging.

Assumptions

  • You have access to an OpenStack deployment.

System Requirements

For Evaluation and Enterprise VMR editions, the system resources that you provision for the VM automatically determines the number of client connections to the VMR that can be made. If you provision:

  • 30 GB of disk space, 4 GB of RAM, and two vCPUs—a maximum of 1,000 client connections are possible. This is the minimum required system resources.
  • 30 GB of disk space, 12 GB of RAM, and four vCPUs or greater—a maximum of 10,000 client connections are possible.

For Community VMR editions, a maximum of 100 client connections is permitted. Increasing the system requirements will not increase the number of available client connections.

Note:  To deploy VMRs in high-availability (HA) redundancy groups, you must set up three separate VMR instances as discussed in this section, and then configure them appropriately as a group. For more information on how to configure existing VMRs as an HA group, see Managing VMR Redundancy.

Download the Solace VMR

The first requirement is to download an edition of the VMR. For Openstack, the VMR is distributed as a Compressed QCOW2.

  • Evaluation or Community edition of the VMR—Go to dev.solace.com/downloads/, then in the VMR Community Edition or VMR Evaluation Edition areas, select Cloud Images, then select OpenStack (Cloud QCOW2). After you read and agree to the license agreement, a compressed QCOW2 file called soltr-<version>-vmr-<type>.qcow2.xz will be downloaded.
  • Enterprise version of the VMR—If you have purchased an Enterprise version of the VMR, Solace will provide information for how to download an enterprise version of the VMR QCOW2 file package from a secure Solace server. Contact Solace Support at support@solace.com if you require assistance.

Once you have the compressed QCOW2 package, you need to extract it so that you are left with the desired QCOW2 image. The compressed QCOW2 can be extracted using the tar command.
Extract qcow2 image
Then you are left with a file called soltr-<version>-vmr-cloud-<type>.qcow2. This is the file you will use going forward with OpenStack.
Note: Some OpenStack implementations claim to support direct qcow2.tar.gzfiles imports. However, Solace strongly recommends importing a decompressed image.

Import the QCOW2 into OpenStack

Note: Your OpenStack GUI may be slightly different than shown below. The following procedure may need to be adjusted to work in your OpenStack environment.

  1. In the OpenStack GUI under Images, click Create Image.
    Create OpenStack Image
  2. Enter details for your VMR Image, then click Create Image. The creation of the image may take several minutes to complete.

Starting the VMR Instance in OpenStack

  1. Prior to launching an instance base on the newly created image it is a good idea to configure a security group to be used by the instance.

    Go to Access & Security and select Create Security Group. When prompted, provide a name and description, then select Create Security Group.The new security group will appear in the list of configured security groups.

    The next step is to an appropriate security rule for each port that the VMR uses for a service to enable connectivity to your VMR; Select Manage Rules.

    For information on the default ports the VMR uses, refer to VMR Configuration Defaults. The example below includes rules for all service ports that the VMR may use. Alternatively you may only expose the services required for your application.

    Note:   

    • Consult your local security policy before configuring access rules.
    • If you will be using the VMR in an HA redundancy group with other VMRs, you must create security rules for ports 5404, 5405, and 8741.

    Security Group Rules

  2. To launch the instance, go to Images, and click Launch Instance.
  3. On the Details tab, specify an Instance Name and Flavor; under Instance Boot Source, select Boot from Image; and under Image Name, select the VMR image that was created in the previous section.

    Note:  The VMR will inherit the Instance Name as its hostname.

    OpenStack Instance Details

  4. On the Access & Security tab, select or create a Key Pair, and assign a security group.
    OpenStack Access and Security
  5. On the Networking tab, assign one and only one network interface to the VMR.
    OpenStack Networking Properties
  6. Click the Launch button. The VMR will launch and the OpenStack dashboard will display the running instance. Here you can find the IP address of the instance.
    OpenStack Dashboard
  7. To log into the Linux Host shell, enter the following command:

    ssh -p 2222 -i <auth_key> sysadmin@<public_ip>

Configuring Users

After initially booting a cloud instance of a VMR, only key-based login for the sysadmin account is allowed. For more information on login and user configuration, refer to Configuring Users in the Cloud.

Managing the VMR Configuration

Unlike a Solace messaging appliance, a VMR starts with a basic configuration that enables most common services. This basic configuration can be modified as required. For more information, refer to VMR Configuration Defaults.