Changing deployment profiles on VMware

You can change an API Connect installation to use a different deployment profile.

For information on the available deployment profiles, see Planning your deployment topology and profiles. The steps for changing deployment profile after installation depend on whether you want to change to a profile that has a different number of VMs and whether additional CPU and memory resources are required for the profile you move to.
Note: The procedures to change deployment profile that are described in this topic do not permit the endpoints to be changed. The API Connect endpoints are listed here: Deployment overview for endpoints. If you want to change a one replica deployment to a three replica deployment, and you used your VM hostnames in your endpoints, then you must configure load-balancers and logical DNS records so that your original VM hostnames route to your new VMs. Alternatively, you can change endpoints by following the form factor migration procedure: Migrating from v10 to v10 on a different form factor.

Moving to a profile that has the same number of VMs

If you are moving to a profile that requires greater CPU and memory resources, follow these steps:
  1. Complete the disaster recovery preparation steps for your subsystems: Preparing for a disaster.
  2. Shutdown your API Connect VMs.
  3. Update the CPU and memory resources for your API Connect VMs in your VMware console, as required for your new profile.
  4. Restart your VMs.
  5. In your project directory, update your deployment profile:
    apicup subsys set <subsystem_name> deployment-profile=<new profile>
  6. Apply the new profile:
    apicup subsys install <subsystem_name>
If you are moving to a profile that requires lower CPU and memory resources, follow these steps:
  1. Complete the disaster recovery preparation steps for your subsystems: Preparing for a disaster.
  2. In your project directory, update your deployment profile:
    apicup subsys set <subsystem_name> deployment-profile=<new profile>
  3. Apply the new profile:
    apicup subsys install <subsystem_name>
  4. Shutdown your API Connect VMs.
  5. Update the CPU and memory resources for your API Connect VMs in your VMware console, as required for your new profile.
  6. Restart your VMs.

If for any reason you are not able to change the CPU or memory on your VMs, redeploy them following the disaster recovery process: Recovering from a disaster.

Moving from a one replica to a three replica deployment profile

  1. Complete the disaster recovery preparation steps for your subsystems: Preparing for a disaster.
    Note: You do not need to take VMware snapshots, since your existing VMs are deleted and replaced when you change deployment profile.
  2. In your project directory, update your deployment profile:
    apicup subsys set <subsystem_name> deployment-profile=<three replica profile>
  3. Define the hostname and network interfaces for the two new VMs in each of your subsystems. Use the apicup hosts create and apicup iface create commands. For more information, see the hosts and interface create steps in the installation instructions for each subsystem:
  4. Shutdown your existing API Connect VMs.
    Note: To minimize downtime, you can delay shutting down your existing deployment until you are about to deploy their replacement VMs in the next step.
  5. Follow the disaster recovery steps for each subsystem: Recovering from a disaster, but note that you must create new ISO files. Do not attempt to reuse the previous ISO file for one of the VMs in your new three replica deployment.

Moving from a three replica to a one replica deployment profile

  1. Complete the disaster recovery preparation steps for your subsystems: Preparing for a disaster.
    Note: You do not need to take VMware snapshots, since your existing VMs are deleted and replaced when you change deployment profile.
  2. In your project directory, update your deployment profile:
    apicup subsys set <subsystem_name> deployment-profile=<one replica profile>
  3. Delete two of your three VMs from your configuration with the apicup hosts delete command.
  4. Shutdown your existing API Connect VMs.
    Note: To minimize downtime, you can delay shutting down your existing deployment until you are about to deploy their replacement VMs in the next step.
  5. Follow the disaster recovery steps for each subsystem: Recovering from a disaster, but note that you must create new ISO files. Do not attempt to reuse any previous ISO files for your new one replica deployment.