Learn how to use an instant clone push image to update a pool of instant clone desktops with a new snapshot.
- [Instructor] In this video, we'll go through some Instant Clone Operations and primarily we're going to focus on how do you update the Master Virtual Machine and how do you push out a new image to all of your Instant Clones in a pool? So first off, updating the Master VM. This is our golden image. We simply power it on and make whatever changes we want. The link between the Master VM and the Instant Clones is indirect. That means that we can update the Master VM however we want without impacting any running Instant Clones. So once we're done making whatever changes we need to, let's say we're doing Windows updates, we shut down the Master, we take a new snapshot of it and then we're going to push an image to the Instant Clone pool. And so this is what we call a Push Image operation. So here you can see our current Instant Clone pool. We haven't made any changes to the Master VM yet. So we got a Master VM. A Template which is a linked clone based on that Master which was used to create this Replica and the Replica is the base disk for all of our parent VMs. So again, running on each EXSi host is one of these parent Virtual Machines. And so the Replica serves as the base disk for these parent VMs. And the parent VMs serve as the base disk for my Instant Clones. They have the contents of memory. They're what allows those Instant Clones to start up so quickly. So that's where things stand right now. But then we decide hey, it's time to make some changes to the Master VM. We shut it down. We do Windows updates for example. We snapshot the Master VM to capture all of those changes that we've just made. And then we schedule out a Push Image. Now a Push Image is very different than a Recompose. They work much differently. But you can think of them as being relatively similar. They're basically going to do the same thing. They're going to serve the same purpose. I've got some sort of change, I want to push it out to all of my Instant Clones. That's when I'm going to do a Push Image. So what I can do is I can go to my Instant Clone pool and I can schedule a new Push Image. I don't have to carry it out right away. But what I can do is I can schedule a Push Image to occur. And so when I push this image, it'll result in the creation of a new Template. And the Template is simply a linked clone of the Master Virtual Machine based on the snapshot that we just took. So I made my changes, I took a snapshot, I scheduled a Push Image and the Push Image resulted in the creation of this new Template. And this linked clone, that's what the Template is, it's a linked clone, will power on temporarily, then it'll shut down. And once that Template is ready, that's when our Replica is created. This is a full copy of the Template that the parent VMs will use as their base disk. So now we've got this Template and Replica up here at the top. This is our original Template and Replica. What our parent VMs are currently based on. We've basically created a replacement down here that our parent VMs are going to be based on after the fact. And that's essentially what happens. New parent VMs are rolled out on all of the ESXi hosts where our Instant Clones run. These new parent Virtual Machines. And then at that point the users can be forced to log off or we can wait until they do it on their own. We can schedule it for a certain time, we can give them a warning. But we've got these new parent VMs and as users are forced to logoff, and log back in, they are getting new Instant Clones. The old Instant Clones are being destroyed. So they're getting entirely new Instant Clones. We're basically blowing away all the old Instant Clones and creating new ones. Now the final thing that I want to mention here, let's go back a slide, when I start a Push Image, I can schedule it. So basically what I'm doing is I'm setting up in the future a Push Image that's going to occur. But if something goes wrong and I decide, you know what, I don't want to push that image. I just realized I forgot something. You can cancel a Push Image before it begins. If it's been scheduled, and it hasn't actually occurred yet, you can cancel it and there's zero impact. If you realize after the fact, let's say that your Push Image has started and half of your Instant Clones have been replaced, you can cancel a Push Image at that point. And the clones that have the new image will remain in the pool and the pool will essentially have a mix of clones with the new image and some of the old image. So what you'll then do is eventually get your ideal image and do another Push Image to replace and destroy all of those linked clones that currently exist so that you can get them at a consistent image. So those are some of the little potential situations that you might run into when you schedule a Push Image. You need to be aware of those when you take your exam. So coming up, we'll take a look at how to create an Instant Clone pool and how to update one by scheduling a Push Image.
- VMware Horizon architecture
- Navigating options within Horizon Administrator
- PCoIP and Blast Extreme display protocols
- Linked-clone desktop pools
- Creating an instant-clone desktop pool
- Delivering Horizon applications
- Configuring roles and permissions
- Using Cloud Pod Architecture