VCF on VxRail: Create vLCM WLD Using WFO Script

Recently I made a post about creating a vLCM enabled WLD using WFO UI. However, sometimes the WFO UI may not have a certain capability or feature yet. In those cases, there is a workflow optimization script that can be used to gather all the information for the new WLD, and then make an API request to SDDC Manager’s backend to start the workflow.

Create vLCM VI Workload Domain

The steps used for the create WLD workflow are the same as prior versions of VCF 5.x, with a few small exceptions. Just like with WFO UI, make sure that all create WLD prerequisites have been completed prior to launching the using the kicking off the WFO script.

Getting Started

Login to SDDC Manager using the administrator account. Navigate to the Developer Center left-pane tab and select Download under VCF on VxRail Workflow Optimization. Once the script has been downloaded, upload the .zip file to SDDC Manager’s home directory /home/vcf/ using a file transfer utility like WinSCP.

Establish an SSH session to SDDC Manager using the vcf user and extract the zip file using by using the unzip command. Change directories into the newly created directory named WorkflowOptimization-VCF-51-master. The python script that will be used in for the rest of the post is vxrail_workflow_optimization_automator.py.

Running The Script

Kick off the script my typing in python vxrail_workflow_optimization_automator.py. The script will now start gathering all input that is required starting with the SSO username and password. In this case, that user account will be administrator@vsphere.local. I’ve attempted to break down the screen captures in categories to make things easier to understand. However, prompted inputs are pretty straight forward. The first set of questions are geared toward the basic WLD settings, deployment type and VC object names. For example, here we are creating a WLD named 37hh-w01 that will join the management domain’s SSO domain using ELM and will be a vSAN based WLD whose datastore name is 37hh-w01-vsan01.

Next the script needs to determine how the user will provide the required input – either VxRail JSON from https://vxrailconfiguration.dell.com/ or a guided step by step. This post uses the step-by-step method to display all the prompts within this particular environment. This new WLD will be using a vSAN OSA (not ESA) cluster with vLCM enabled. Now the vCenter network gateway and subnet mask are required, along with the datacenter name and cluster name.

Here’s the VxRail specific and networking portion of the script. Provide the FQDN for VxRail Manager so that it can be resolved, and thumbprints fetched and excepted. The VxRail host discovery API will be used to identify VxRail hosts discovery by VxRail Manager. Enter in the number beside each host’s PSNT that needs to be used for this VxRail WLD deployment followed by each of their FQDNs when prompted. The password now needs to be input – this can be done with a single password or specific passwords for each host.

Define all of the required networks for the VxRail deployment. vSAN is the principal storage type for this new WLD so that network is required, along with the default vMotion and Management Network (ESXi Hosts and VxRail Manager). In VCF 5.1 and newer WLDs are deployed with ESXi management vmks and “VM Management” (known to VxRail users as the vCenter network) on different port groups. By default, these port groups are the same on the same VLAN. There is a new feature in the script asking if “VM Management” details are needed, which will provide an option to allow these networks to be on different VLANs. In this post, we will not be diving into this feature.

Provide the NIC Profile to be used for the deployment, TWO_HIGH_SPEED was used in this environment, on a shared system/overlay VDS. All the VC object names for VDS name, MTU, port group names are required to be input and can be customized. However, the Management Network must be prefixed with Management Network at a minimum or the deployment will fail. Lastly for the networking section, select the VDS to be used for overlay traffic, which this environment is using shared so there’s only one option and provide the overlay VLAN ID.

Provide the NSX Manager Appliance information. Input the username, password, FQDNs for the virtual IP and three managers, as well as TEP IP deployment method of DHCP or static pool. VxRail Manager users accounts are required as well, so input the root password as well as the mystic password.

Finally, select all the licenses to be used (these are already in the SDDC Manager’s database) and whether or not the vSphere licenses need to be applied now or if they will be done after the deployment.

Now that all the input has been provided, the screen first displays the components of the vLCM cluster image as well as the JSON file that will be sent to the VCF backend for deployment (save this JSON file incase a retry is needed if validation fails). 

Press enter to continue which kicks off the validation and wait for successful validation. If validation is successful press enter again to start the deployment and follow along in SDDC Manager’s UI. If for some reason it fails validation please refer to VCF on VxRail: WFO Script ‘Retry’.

Finishing Up

Now that validation was successful, all status of the WLD deployment will be displayed in SDDC Manger. Notice the new WLD and it’s Configuration Status is set to ACTIVATING. Once completed the status will change to ACTIVE.

It’s easy to verify that vLCM has been enabled on the cluster by navigating to the cluster in vCenter > Configure > Desired State > Image. This is just one of many ways check that vLCM is enabled from within vCenter. It’s also possible to query the VxRail API to determine if vLCM is enabled.

Helpful Links

Leave a Reply

Create a website or blog at WordPress.com

Up ↑

Discover more from vbarneeze.cloud

Subscribe now to keep reading and get access to the full archive.

Continue reading