Nerf Wars!


I was recently taking  a training class across town. During the class the instructor said, “Oh, at about three o’clock the room next to us will have a nerf gun war. So, sorry if they get loud.”

Most of us in the room looked at each like “what?”. Well, right at three o’clock, not only did we hear the nerf gun war but we saw nerf bullets flying by as well.

I have to be honest, this put a smile on my face. What a cool and fun way to let off some steam! I could tell this place was all about fun because they also had a ping pong table in the middle of the floor with plenty of balls and paddles as well as snacks and soft drinks freely available.

Boosting morale is something that companies should have high on their to do list. Happy employees should be a top goal – and is something that gets pushed aside due to “business as usual.” It’s not always expensive and difficult to achieve this. It may just take a few toy guns and a few minutes each day.

Posted in Morale | Tagged , | Leave a comment

Prioritization between projects

The most efficient way to prioritize projects and complete the work is one item at a time. See the below diagram for further explanation:


This diagram was borrowed from the book “Scrum” by Jeff Sutherland (which I highly recommend). The explanation is as follows:

1.) The top line shows that traditional project management follows the line of thinking that everything is important so work on every project at once. By working multiple things at once, you lose a % of time to context switching i.e. waste. For example, working on three projects at once means you lose 40% purely to waste and only leaves 20% per project. (Scrum pg. 91).

2.) The bottom line shows that Agile says to focus only on one project at a time. You do not incur waste due to context switching and you finish all three projects sooner than focusing on all three at once.

Strategy: Focus on as few items as possible. This goes along with WIP limits. As well, it may be necessary to enforce hard project limits per person i.e. no more than two projects at any given time.

Posted in Management | Tagged , , | Leave a comment

Team Health Check

This is an idea that leaders can use to gauge the overall health of a team. This should NOT be used in a negative way such as comparing one team to another or making a team feel bad for having a lower status. The point should be that if a team rates themselves low then we must ask why. Here is a very simple graph that can track this metric:


The team should rate themselves and no one else. If a team is in the red we should start asking questions. Why did they rate themselves as red? What can we do as servant-leaders to improve their status? They should put reasons next to why they rated themselves a certain way.

Strategy: This is a drawing that can be made on a marker board or a stick and peel easel pad. It can then be placed into an area that the team often frequents such as the daily stand up. Let the team choose their health status and the reasons why. Use real sticky notes as shown above. There should be consensus from the entire team when going from one status color to another. This can be used for both scrum and silo’d teams alike.

Posted in Management | Tagged , , , | Leave a comment

Leadership at all levels

If you’re not being a leader because of your title then you’re missing out. I’d like to think that I’m a leader even though I’ve never been a manager or executive and haven’t had an official “leader title”. The point is that your level, be it a new hire or seasoned veteran, should not dictate whether you bring leadership to the table.

Leadership is one of the most important intangible skills because it takes a combination of the others (communication, positivity, etc.) to possess it. During hiring I look for leadership skills and other intangibles over technical skills because, in my opinion, it’s harder for someone to learn those skills vs the “0’s and 1’s.”

The book The World Is Flat is about economics, but you could also apply that title to leadership. Businesses are increasingly going to a flat management style. This is putting more onus on individual contributors than ever before. I think this is  a good thing. Too many layers creates bureaucracy. It also means, like it or not, you’re going to have to step up your leadership game. Better to start now than to wait around.

Here are some ideas on things you can do to further your leadership skills:

  • ALWAYS have a positive attitude. I can’t stress this enough and it’s personally my #1 leadership skill. I’ve been around too many people with jaded attitudes and it’s just awful.
  • Do good work. It’s that simple. If you’re putting out the best work and are the most efficient team member others will recognize that. It may also lead to promotions.
  • Have answers to questions that no one else does. This is not easy but it is possible. If you’re always the one with answers management will go to you above anyone else.
  • Think outside the box. This is a bit of  a cliche, I know, but it is true. If you do things only because that’s the way they’ve always been done then you’re not innovating.
  • Innovate. You can do this by learning a new skill-set or bringing new ideas to the table.
  • Lean on someone you feel is a leader and learn from their habits.

Update 3/30/16: I came across an article on linked in called How TO Be A Leader When You’re Not The Boss. It was written by Ilya Pozin, the founder of Pluto TV and  writer for Forbes. I was extremely surprised by this article for two reasons: One, it was a nearly identical article to this post, and two, in the article is a survey done by Waggl that says the top Human Capital priority of 2016 is “Leadership at all levels” – the exact title of this post. My post was written in Oct. 2015. The survey was done in Dec 2015 and Ilya’s article written 3/2016. It’s gratifying to know that my thinking is in line with Ilya’s! 🙂

Posted in Management | Tagged | Leave a comment

Create vCO/vRO scheduled workflow automatically

With vCAC/vRA, when you “destroy” a machine it deletes the vm immediately. There is a setting you can disable that will not delete it but instead will power if off and send it to a default “destroyed” folder in vCenter. However, after that it just sits there until it’s manually deleted.

Due to a business need, I wrote a workflow that automatically deletes the vm after a given number of days. In my example, we delete vm’s 14 days after they’ve been shut off. You can edit the code to make it however many days you’d like though.

First, if you haven’t changed the destroy option yet, check out this blog on how to do that. Once done, go on to the next step.

Once you have disabled the immediate delete feature, you’ll want to do the following in your vCO/vRO console:

In your Decommisson workflow, drag and drop a scriptable task element on the page. You’ll probably want to make it one of the very last elements. Here are the properties:

and the script is:

workflowScheduleDate = new Date();
var day = workflowScheduleDate.getDay();
workflowScheduleDate.setDate(workflowScheduleDate.getDate()+ (day == 0 ? 14 : day == 1 ? 14 : day == 2 ? 14 : day == 3 ? 14 : day == 4 ? 14 : day == 5 ? 14 : day == 6 ? 14 : 0 ));

//This workflow ID is Delete virtual machine — copy and change to suit other workflows
var workflowToLaunch = Server.getWorkflowWithId(“BD80808080808080808080808080808003C180800122528313869552e41805bb1”);
if (workflowToLaunch == null) {
throw “Workflow not found”;

var workflowParameters = new Properties();
var scheduledTask = workflowToLaunch.schedule(workflowParameters,workflowScheduleDate);


The most important part is understanding how the days are calculated. If you see where it says “day ==0 ? 14”, this is javascript saying if day = Sunday, then schedule the deletion 14 days later. You could work the numbers to be business days only. You’d have to take into account the number of days and weekends etc. Our policy is 14 days from the time of shut down – business or not. Nevertheless, if your policy is 30 days, then just replace the 14 with 30.

Update1: If you want to give the user the ability to choose the number of days, all you have to do is create an input parameter type number called something like “numDays” and replace all of the “14’s” with the numDays variable. Then, when the workflow is run, you can put in any numeral and the workflow will be kicked off that number of days later.

Lastly, be sure to change the long character string that is the delete vm workflow. In essence, you could schedule any workflow with this script – you just have to change the workflow ID. To get the workflow ID, click and highlight the given workflow, then hit ctrl-c, then open notepad and ctrl-v. This will paste the workflow ID!

Now when you “destroy” a machine in vRA, it will be shut off, moved to the default folder, and deleted 14 (or however many) days later!

Update2: If for whatever reason you want to kill the deletion of vm, all you have to do is the following: Open vCO/vRO client -> From the main “Run” page and on the home tabe, click on “Tasks scheduled in the sytem.” Find the scheduled task created for your vm and delete it. That’s it!

Posted in Uncategorized | 1 Comment

vCAC 6.0.x and vCO to rename a virtual machine

This post is to demonstrate how to use vCO with vCAC extensibility to rename a VM during vCAC 6.0.x provisioning automatically including an incrementing unique identifier.

I’ve read many blogs and posts on how to do this. None of them show how to do it without vCAC Designer, powershell, or both. This “how to” utilizes ONLY vCAC and vCO (no designer or powershell needed). From what I have read, the focus moving forward will be on vCO with Designer being deprecated. So, if you don’t know vCO or don’t use it with vCAC I would learn it.

First, I’m using vCAC I’m not sure the minimum version needed but 6.0 would probably work. I’m also using the latest vCO 5.5.1 appliance. The prerequisites you need are the vCAC 6.0.1 plugin . This will give you all of the workflows that you need in vCO. As of this writing, the link to the plugin is:; title=”vCAC 6.01 vCO plugin”></

For this to all work, you must set up vCAC extensibility with vCO. There is a really good blog post on how to do this. The only difference is you will not update design center but instead the settings will be tied between vCAC and vCO directly. Just be sure your vCO endpoint is setup on your vCAC. I would recommend reading part 2 for a complete understanding of how this works.

Once you have this piece complete and your External stubs installed (WFStubBuildingMachine,MachineProvisioned, etc) you must then set it up so that when you kick off a VM build it runs whichever stubs you want it to. Typically, the standard is to run WFStubBuildingMachine for things like renaming and setting other custom properties i.e. IP ,storage, etc., then WFStubProvisioned for custom installs after the vm has been built. An example would be adding the machine to DNS (if linux) or joining it to the domain or installing components like IIS. Lastly, the WFStubMachineDisposing is used for the disposal of the machine. So, things like removing it from AD, unlocking the name if you use the System locking method etc.


To do this you will use the following workflow: “Assign a state change workflow to a blueprint and its virtual machines.” The way to do this is the following: when you run it, one of the steps it asks is to assign a workflow to be run during one of the states (building, provisioning, disposing). You will want to create three separate master workflows: one for each state (Building, Provisioning, Disposing). Depending on the master workflow, you will build around it based upon the functionality (see above as to why you would use each one).


Select the first option, which is BuildingMachine. Set your master “BuildingMachine” workflow as the workflow that it calls and select the various blueprints you want to kick the workflow off. From now on, anytime you build a machine from that blueprint, it will now call this master workflow that you selected. Do the same for Provisioning and Disposing. If you ever want to remove these workflows you can just run the “remove a state change workflow from a blueprint and its virtual machines” or manually delete them in your blueprint in vCAC itself:


In my example, I have set my master workflow as “Building Workflow”, “Decommission workflow”, and “Provisioning workflow.” To rename the VM to something dynamic, we will be using the Building Machine Workflow.


Here is what my workflow looks like:


To explain this a bit, we have the following needs for a VM name:

Domain makes up the first 3 characters; Location the next three; app code the next three; If it’s a web or database server we add an additional character; we set the classification (Production, QA, Test, etc); Generate a unique ID (01,02,03) then update the properties so vCAC uses the new name.

To keep this example simple, let’s just use two pieces. Let’s say we want a Domain and app code. To set this up, we will want to create a new property in vCAC. Select one of your blueprints, go to the properties tab, and create a new property called “Custom.Domain” with nothing in the value field. Create another called “Custom.AppName” with nothing in the value field.


The best way to set this up is to use a property dictionary. Here is a good blog post on how to do that. Just mimic this for what we’re doing.

Here are some screencaps of my property dictionary. You can probably just copy what I’ve done since the blog post goes into a bit more detail than we need for this. Be sure and name your property dictionary entries the same as what you called the properties you created in the blueprint (this is what links them together). Of course, you can always put static values in the “value” field of your property if you just want to test.


Click on edit under “Property attributes”. Select type “valuelist” and enter your values separated by a comma. This is what will be in the dropdown of your blueprint.

At this point, we can begin to manipulate the actual workflow. Drag and drop a scriptable task onto the form. This will be to grab the Domain name. Here are my inputs, outputs, and script:




Drag another scriptable task onto the pane after the first. This one is for the App Code.




(the second line says to capture from the first character until the first space. My app codes are setup in the following way “Code – Name”. So, you can remove this line assuming you just put in app codes as “SQL”, “ORA” etc.)

At this point our serverName variable is composed of domain + app code. If our domain was “TST” and our appcode “ABC”, the variable would be TSTABC.

Possibly the best part of this for some will be the creation of the unique ID i.e. a unique number at the end of the VM. After all, if you have 1000+ vm’s, it can be a major pain to come up with a unique name automatically. I will include the code here and explain it:


(On a side note, I am not a programmer and I did not write all of this code. I had a basis including the search functionality and had a programmer friend tweak it to make it more efficient)

Copy and paste this into a new scriptable task. The one input is serverName and the one output is finalName (string/attribute).

Here is what it does: It searches both vCenter via the vCenter plugin (pre-requisite for this to work) and the internal vCO locking system to see if the name already exists. If true, increment the number by 1. For example, if TXTABC01 exists, then it will make the name TXTABC02 etc. It also locks the new name in vCO from being used again (be sure to set an unlock workflow to run during your decommission stage or remove the locking line towards the end all together).

Drag and drop a third scriptable task onto the pane, and call it something like “Update Properties.” Here are my ins/outs/script:




Finally, drag and drop the workflow called “Updated a vCAC model entity.” Here are its properties:



That’s it: Once the workflow is finalized, your vm will have it’s new name. Be sure you have your customization spec set in your blueprint from vCenter or else sysprep won’t change the name (it will show up in vCenter correctly but the actual OS won’t reflect the change). You’ll set the option that says “Use the virtual machine name.”

Posted in vCAC | Tagged , , , , , | 6 Comments