In the previous post we covered the architecture and setup of the vRealize Code Stream Management Pack for IT DevOps (also known as Houdini). In this post we’ll cover how we need to setup Houdini’s endpoints so that we can use them to release our blueprints or workflows to other instances.
Remote Content Server Endpoint Setup
To setup our endpoints we can use nicely packaged blueprints right in vRA. It’s pretty nice that our setup deployed some blueprints for us to use, right in the default tenant of our vRA server. Login to the vRA default tenant with your Houdini Administrator that you setup in part 1. Then go to the catalog and request the “Add Remote Content Endpoint” under the “Administration” service. A remote content server (RCS) is a vRA appliance that will cache your packages. It’s a pretty useful thing to have if you’ve got vRA appliances in different sites and you need to move vSphere VMs or other large objects over a WAN. Future releases can be copied from the remote content server instead of always copying from the source.
The first screen of the blueprint will show you information about the request. Read through it and then move on to the “Add Remote Content Server” tab. In the Remote Content Server Name field, give it a name. Then enter in the connection information for the content server.
Next, vRA will do some verification checks before you submit the request. Check the SSH fingerprint and accept it. If an error is occurring, go back to the previous step and check your connection information.
On the last screen, you can have the request data emailed to you and you can do a “mock request” which won’t really do anything other than test the request outcome to see if it would work. I’ve skipped those steps in my screenshot. Click Submit and you can watch the request happen in the requests tabs of vRA.
Setup the vRealize Orchestrator/Automation Endpoints
Now we need to setup our development server endpoints. These endpoints will be considered our “source” endpoint which is where our blueprints are built and then captured by Houdini. The endpoint setup requires us to add endpoints for both vRealize Orchestrator and vRealize Automation, IN THAT ORDER! This is done in the same manner that the remote content server endpoint is setup. This time go to the catalog and request the “Add Content Endpoint” blueprint in the “Endpoints” service.
Again, the first tab is the information tab which you can read through, when you’re ready, go to the Add Content Endpoint tab. Give the endpoint a name and select the Orchestration category. Under the supported package types select all of them and click the arrows to move them into the box on the right side that isn’t really labeled. After that you’ll want to enter the FQDN of your development vRealize Orchestrator instance (which may be embedded in your vRA appliance). Then enter connection information about your vRO instance and you can assign tags for describing the instance.
On the “Endpoint Policy” tab, select “Allow capturing from this endpoint” check box. You could select all of the check boxes but in my example this is a one way operation, meaning that I’ll capture objects from this endpoint and release them to another endpoint. This setup won’t support the opposite method in my configuration but your situation may be different so check the boxes that make sense in your environment. You’ll also notice that you can configure a vRA tenant for testing your blueprints before releasing if you’d like. The tests are run from vRealize Orchestrator and you’ll need to build some of your own tests for it to be truly useful.
On the “Connection Test” tab, click the “Test connection” check box to ensure that the endpoint is responding. When the certificate fingerprint comes back, verify that it’s correct and then accept the certificate.
In the “Submit Request” tab we have the same options as before. Do as you like here and then click submit.
After you’ve configured the vRealize Orchestrator Endpoint, you’ll do the exact same operation for a vRealize Automation Endpoint. Run the same endpoint blueprint and fill out the endpoint name. This time in the content category drop down select “Automation” from the list. After that you’ll need to select the vRA version for the development vRA instance. Again select whichever vRA packages that you’ll want to manage and push them to the right hand box. Under the attached vRO servers, select the endpoint we setup in the previous steps. Then enter the connection information for the development vRA instance and the tenant which you’ll manage.
Again, we’ll select the “Allow capturing from this endpoint” check box since we’ll export blueprints out of this instance and not into it.
Test the connection again, just like before and accept the certificate fingerprint. Submit the request.
Add vRealize Orchestration/Automation Production Endpoints
Perform the same steps as above again. Setup an orchestrator endpoint and an automation endpoint for the production instance. The process for this is exactly the same except on the “Endpoint Policy” tab we need to select “Allow releasing to this endpoint” instead of the capturing from this endpoint check box. You can also select our remote content server which is optional for your setup. I’ve also gone through this setup again and added a testing endpoint which is just an additional tenant on my development instance. You could use an entirely different vRA instance or whatever you think would work best for your situation.
If you’ve made it this far, you’re ready to start using Houdini to move your blueprints around between instances. In the next post we’ll do just that by leveraging the vRealize Code Stream Management Pack for IT DevOps.