Citrix released Citrix Workspace Cloud (CWC) earlier this week. Most people I talk to agree that CWC is one of the most innovative solutions in the industry right now. You can read more about my thoughts on CWC here.
In this post, I want to focus on one of the components of CWC namely Citrix Lifecycle Management (CLM). CLM provides customers and partners alike an orchestration and automation engine that they can use to build an end to end infrastructure that meets industry and their own best practices on ANY cloud.It also helps you manage and monitor the environment as well as conduct operational tasks and upgrades.
What is truly great about this solution is that CLM is not limited to deploying Citrix infrastructure. It can be used to deploy Microsoft infrastructure (AD, SQL etc), partner solutions (Atlantis USX, LoginVSI, Xangati etc) in addition to Citrix infrastructure (XenApp/XenDesktop/Netscaler/Storefront etc). There are various blueprints that are already pre populated and the list will continue to grow. In addition customers and partners can create their own blueprints based on their own best practice. Because blueprints are leveraged, deployments are easily reproducible and error free. For instance, a senior architect could develop a complex blueprint and then once complete, hand off to a junior resource who could very easily deploy the same blueprint anywhere he pleases with consistent results. From a partner perspective, this makes deployment of solutions extremely simple and consulting engagements and proof of concepts should end up being much more successful and predictable.
In the next part of the blog, I would like to walk through an on premises deployment of an on premises XenDesktop POC that I built leveraging CLM.
Once you log into CLM. The first thing you want to do is to go ahead and define your resource zones. Today CLM supports the resource zones listed below. In my case, I leveraged an on premises XenServer 6.5 host.
The next step in the process is to install the CLM connector on Windows Server 2012 R2 server. This is similar to a CWC deployment. The connector is what facilitates communication between the CLM orchestration engine and your resource zone. Once the connector is installed, you define the connection details for your host and you are all set. As you can see from the screenshots below, I defined two resource zones in my lab, both XenServers.
Now that we have our resource zones, the next step is to go through the blueprint catalog and pick the one we want to deploy. As mentioned earlier, there is a pretty good list of blueprints already available and this list will continue to grow over time. Once you know the blueprint you’d like to use, you click the “+” symbol to add the blueprint to your library.
For the purposes of my test, I used the XenApp and XenDesktop Proof of Concept Blueprint and added it to my library. The blueprints that I choose will show up under the “Design and Deploy” tab. Within “Design and Deploy” you now click on “Actions” next to the selected blueprint and choose “Deploy”. Note that you have the ability to edit, clone or share the blueprint if you want to. This will allow you to customize the blueprint as you please.
Once you hit “Deploy”, you are provided a description of the blueprint along with the various infrastructure components. After reviewing the details, click “Start Deployment Setup” and give the deployment a name. You can also choose an existing deployment profile if you’ve gone through this process before.
As part of the template I chose, I have some additional options like providing a KMS Server address, configure a Netscaler, create a server VDI instance etc. I have stuck to the defaults for the test. I also specified the resource zone that I would like the solution to be deployed on.
Next, you’ll see a pre deployment checklist based on the resource zone you selected. Review this carefully!! One of the key pieces to having this work properly is to make sure you have a Windows Server 2012 R2 template ready on your XS host. Below are some things to keep in mind:
- Make sure the image is sysprepped with the /generalize /shutdown /unattend options. Use the unattend file to provide login credentials among other things. A sample unattend file can be found here.
- Make sure RDP access is enabled in the template
- Make sure the Windows Firewall allows outbound connections on port 443
- Install the latest XenServer tools and reboot the machine as many times as needed to ensure the process is complete.
On the next screen, hit edit on each of the VMs and define the various parameters. Some of the key things you need to make sure include selecting the right VM template, NIC and providing the correct credentials. I have included screenshots for one of the VMs below.
After going through the configuration for each of the VM’s hit “Next” and provide additional configuration parameters in relation to the domain, XenDesktop and computer names. You can also import a config file to fill these parameters. Once complete, give the deployment profile a name and description.
On the next screen, review the configuration and hit “Deploy”. This starts the deployment process. If you look at your hypervisor host you will notice that VM’s being to get created.
Next click on the “Manage” tab and click on your current deployment to check the progress of the deployment. You will obtain a detailed breakdown of each step of the process.
You will also be notified via email as each VM is successfully created and also when the whole deployment is complete.
The entire process in my case took about 2 hours on really old hardware. I will be redoing this test in the coming weeks on much faster SSD drives to see how much time I can shave off.
In addition to allowing you to deploy solutions, CLM also has inbuilt monitoring and alerting. You get a snapshot of how the VMs are performing and you can also set thresholds for the alerts.
CLM also allows you to define operational tasks such as installing updates, running certain scripts etc right from within the CLM engine. These tasks can also be scheduled.
You can also define rules for dynamically tearing down, scaling up and also failover/fallback rules with the CLM engine. In the future I plan to play around with these a little more.
I am truly excited about the potential this platform holds and the value it brings to customers and partners. CLM helps automate the deployment, manage and monitor the environment and tear down whether you want to deploy just within your datacenter, a public cloud, private cloud or a hybrid cloud. In other words it helps through the entire lifecycle of a deployment. The simplicity of the platform along with the blueprints ensures that you dont need to have expert level skills on any of the solutions to deploy the blueprints. I would encourage everyone to try it out. I am sure if you do, you will end up being just as impressed as I am today!