The first step to getting this process to work is to realize that ultimately under the hood, vCO is a Linux appliance, with a lot of functionality not exposed, or not immediately obvious. There are a lot of tweaks you can make to really enable great functionality, and this process may give you other interesting ideas!
NOTE: The below steps assume vCO Appliance 5.5+
You’ll need to start out by using PuTTY to SSH into the appliance. If SSH is turned off, you’ll either need to use the console, or enable Administrator SSH from the VAMI interface.
Once logged in, change directory to /etc/vco/app-server, and then type vi vmo.properties to open the file in a text editor.
Inside you will want to see if you have a line that looks like this:
If it doesn’t exist, press i to change to Insert mode, and then add a new line and put it in there. Once done, press ESC to exit Insert mode, and type :wq to write the file and quit the editor.
When vCO Server is restarted, you will be able to execute Linux commands against the VM from within your workflows. The catch is, you have to make sure that the vco account on the appliance has the ability to execute it.
To enable this, type vi js-io-rights.conf from the shell. This file may already exist and have some data in it. If not, you get to define explicit rights at this point. Here’s mine for reference:
Add the below lines to the file by pressing i, again going to Insert Mode and then the below information, with each line corresponding to a specific executable on the appliance. The prefix characters are adding read and execute rights for the vco user.
Press ESC, then :wq to save the file and exit.
With these tweaks enabled, you will need to restart vCO Server. You can do this a number of ways, but since you’re in the shell this is the fastest:
service vco-server restart
Now, you will be able to execute these commands in a workflow when you use the Command scripting object, which will run the command and return the standard output back as a variable, as well as the execution result, such as success or failure!
With that in mind, let’s do a quick experiment in a workflow to ensure it works as intended.
Proof of Concept Workflow Creation
- Create a new Workflow as you normally would. No inputs are necessarily required for this test as we will get into those values in later posts.
- Drag and drop a fresh Scriptable Task into the schema, and edit it.
- Paste the code below into the scripting tab:
// Creates new Command object, with the command to run as your argument var myCommand = new Command("/usr/bin/openssl version") // Executes the command myCommand.execute(true) // Outputs the result code, and the output of the command System.log("Exit Code: "+myCommand.result) System.log("Output: "+myCommand.output)
Close the Scriptable Task, run the workflow and check the log – you should see something like this:
If you were to type the same command in the shell, the result would be the same. So while we’re here, let’s update the code in the task to verify cURL also works. Change line 2 in the task to look like this (note the case-sensitive argument!) :
var myCommand = new Command("/usr/bin/curl -V")
You’ll probably note that the OpenSSL version installed on the VCO appliance is the same one that is required by VMware for the entire SSL implementation in the 5.x release! With this working, now we can do some really cool stuff.
In the next post, we will build out the workflow that will create the private key and CSR for any and all of your ESXi hosts! This same flow can be used as the basis for vCenter SSL, vROPS, or even vCO itself!