How to set full CPU and memory Reservations for Platform and Collector nodes in our Clustered Deployment of VCF Operations for Networks

Products

• VCF Operations for Networks

Issue/Introduction

Per VCF Operations for Networks documentation at System Recommendations and Requirements, the reservation for the CPU speed and RAM for each node must be 100% of the value specified in the Recommendations for the Platform Deployment

 

NOTE:  VCF Operations for Networks was formerly named Aria Operations for Networks (AON), and prior to that was named vRealize Network Insight (vRNI).

Environment

VCF Operations for Networks

Resolution

1) First, the setting of the reservations will need to be done with all of the platform and collector node VMs in a powered off state.  

 

2) If you have a clustered deployment (i.e. more than one platform node), the powering down of the environment MUST be done using the KB Best practices to shutdown VCF Operations for Networks Clustered deployments. 

• NOTE:  If you have a simple deployment (i.e. one and only one platform node), then you can power off the platform and collector VM(s) using the vSphere client and the Power --> Shut Down Guest O/S action.

• For the clustered deployment, Steps 1, 2 and 3 of that KB Best practices to shutdown VCF Operations for Networks Clustered deployments should be followed in sequence. 

• Step 4 of that KB does not require all of the sub-steps for this reservation-setting operation, because the ones relating to VM snapshots will not be needed for this operation.

  • Sub-steps "a." through "h." must be done to get all of the platform nodes into a powered off state.

  • Sub-step "i." can be skipped, because it relates to VM snapshots, which will not be needed for this operation.

  • Before proceeding with sub-step "j." of that KB Best practices to shutdown VCF Operations for Networks Clustered deployments, the CPU and Memory Reservations settings must be changed, using the following steps:

 

3) CPU: (Do the steps below for each of the platform and collector node VMs)

1. Locate the VM: In the vSphere Client (web or desktop), find your virtual machine in the inventory.
   
   
2. Edit Settings: Right-click the VM and select Edit Settings.
   
   
3. Navigate to Resources: Click the Resources tab in the settings window.
   
   
4. Select CPU: In the "Settings" area (or "Virtual Hardware" tab), click on CPU.
   
   
5. Set the Reservation:
   
   
   • Find the Resource Allocation section.
     
     
   • Use the slider or input field for Reservation to set it to 100% or the maximum available MHz/GHz for your VM's configured CPU cores.
     
     
   • The reservation guarantees this amount of CPU power, expressed in MHz (or GHz), preventing overcommitment and ensuring it's available.
     
     
6. Apply Changes: Click OK to save your settings. 

 

4) MEMORY:  (Do the steps below for each of the 15 platform and 8 collector node VMs)

1. Power off the VM in the vSphere Client. Right-click the virtual machine in the inventory, and then select Power > Power Off.
   
   
2. Edit Settings. Once the VM is powered off, right-click it again and select Edit Settings.
   
   
3. Navigate to Memory Settings. In the Virtual Machine Properties window, select the Resources tab, and then select Memory. (In newer vSphere clients, you may find this under the Virtual Hardware tab by expanding the Memory options).
   
   
4. Set the Reservation. In the Resource Allocation panel, you have two options to achieve a full reservation:
   
   
   • Select the check box labeled Reserve all guest memory (All locked). This automatically sets the reservation value to match the total configured memory size of the VM.
     
     
   • Save the changes. Click OK to apply the new memory settings.
     
     
5. Power on the VM. Right-click the VM and select Power > Power On. 

The VM is now guaranteed to have dedicated access to its full allocated memory from the host's physical RAM. If the ESXi host does not have enough unreserved physical RAM to meet this reservation, the virtual machine will not power on. 

5) After all of the steps in 3) and 4) above are complete for each of the 15 Platform and 8 Collector nodes, resume with Step 4, sub-step "j." of the KB Best practices to shutdown VCF Operations for Networks Clustered deployments

• Perform Step 4 of the KB, sub-steps "j." through "n." in sequence, one sub-step at a time
  
  
• Step 4, Sub-step "n." is the final step in this operation.
  
  
• NOTE:  For a simple deployment (not clustered), you can power on the Collector node VM(s) first, then the Platform node.

 

6) If, after you have completed Step 4, Sub-step "n.", you see anything you feel is an issue, please open a case with the VCF Operations for Networks team using the instructions at Creating and managing Broadcom support request (SR) cases, describing the symptoms of what you see which you feel might be an issue.