You can customize the kubernetes cluster with custom infrastructure requirements like custom packages, network adapters, kernels, and so on, using infrastructure designer. These customizations are available only for CNF components.

You can use the VMware Telco Cloud Automation to customize the infrastructure requirements of the node pools. You can define these customizations through user interface and the system adds these customizations to the corresponding TOSCA file. For more details on the TOSCA components, see TOSCA Components.

From the VMware Telco Cloud Automation 3.1 release onwards, on the Infrastructure Designer page, you can specify configurations for either the Photon OS version 5 or 3 as default. However, if you want to specify configurations for Photon OS versions 5 and 3 both, then you can add the configuration for Photon OS version 3 by clicking + ADD CONDITION button.

Procedure

  1. Log in to the VMware Telco Cloud Automation web interface.
  2. Select Catalog > Network Function.
  3. Click Onboard on the Network Function Catalog page.
  4. Select Design Network Function Descriptor on the Onboard Network Function page. Add the following details:
    • Name: Name of the network package.
    • Tags: Associated tags for the network package. Select the key and value from the drop-down menu.
    • Network Function: Select the type of network function. For infrastructure designer, select Cloud Native Network Function.
  5. Click Design.
  6. On the Network Function Designer page, click Infrastructure Requirements.
  7. To design the infrastructure, enable Configure Infra Requirements.

    • Network Adapter - Click Add to add a new network adapter. Enter the following details:

      • Device Type - Select a device type from the drop-down menu.
      • Network Name - Enter the name of the network.
      • Resource Name - Enter the name of the resource.
      • (Optional) Target Driver - Select the value from the drop-down menu.
      • Interface Name - Name of the interface for the vmxnet3 device. This property is displayed when you select vmxnet3 in Device Type.
      • (Optional) Count - Enter the number of adapters.
      • PF Group - Enter the name of the PF group for which you want to add the network adaptor.
      • Shared Across NUMA - Select the button to enable or disable sharing of the devices across NUMA.
        Note: Shared Across NUMA is appliable only when NUMA Alignments is enabled.

      • Additional Properties - This property is displayed when you select vmxnet3 in Device Type.

        • CTX Per Dev - To configure the Multiple Context functionality for vNIC traffic managed through Enhanced Datapath mode, select the value from the drop-down menu. For more details, see CTX Per Dev. For more details on Enhanced Datapath settings, see Configuration to Support Enhanced Data Path Support.

        • Key-Value List – To configure the Receive Side Scaling function for supported vmxnet3 device type.

          • ctxPerDev – Receive Side Scaling need to set 3, which indicates that multiple contexts (multiple logical cores) are enabled to process each vNIC.

          Note:

          This is the replacement for the legacy CTX Per Dev described above. Unselect the legacy CTX Per Dev configuration if already configured and re-select here.

          • pnicFeatures – Receive Side Scaling need to set 4, which indicates RSS features are requested by vNICs

          • rssOffload – Receive Side Scaling need to set "true", which configure a dedicated RSS engine to process requests from a vNIC

          • udpRss – Set 1 if want to enable the Receive Side Scaling function for UDP traffic

            For more details about Receive Side Scaling, see

            Receive Side Scaling.

      • Sub-Interfaces – This property is displayed when you select vmxnet3 in Device Type. Click on Add Sub Interface to add vlan sub-interfaces and specify meaningful input labels. These input labels will be used at the time of Network Function instantiation to accept the vlan ID for the corresponding sub interfaces.
      • Network Configurations – This property is displayed when you select vmxnet3 in Device Type. Click on Add Configuration to configure vmxnet3 network configuration as the OS level.
        1. Configuration File Type: Select extension of the text file that encodes network configuration for matching network interfaces.
        2. Select Variable: Select the network configuration to add.
        3. Select value for the selected network configuration.
    Note:

    When you select Target Driver, the system automatically adds dependent custom packages in the Custom Packages.

    • PCI Pass Through - Click Add to enter the PTP or PCI Devices.
      Note:
      • A total of 128 SRIOV networking adapters and PCI devices can be configured for a CSAR.

      • When you add a PCI Pass Through device, the system automatically adds the required Linux-rt in Kernel Type and dependent custom packages in the Custom Packages.

      • For the PTP devices, add the following information.
        Note:
        • To use the PTP PHC services, enable PCI passthrough on PF0 on ESXi server when the E810 card is configured with multiple PF groups.
        • To use the PTP VF services, disable the PCI passthrough on PF0 and enable the SRIOV on both the PFs. E810 card supports 1 VF as PTP and the other VF serves as SRIOV VF NICs for network traffic.
        1. Device Type - You can select to add a PTP device or a NIC device. To use a physical device, select NIC. To use a virtual device, select PTP from the drop-down menu.
          Note: To upgrade the device type from PTP PF to PTP VF, delete the existing PTP PF device and add the new PTP VF device. Do not change the device type from NIC to PTP directly in the CSAR file.
        2. Shared Across NUMA - Select the button to enable or disable sharing of the devices across NUMA.
          Note: Shared Across NUMA is applicable only when NUMA Alignments is enabled.
        3. PF Group - Enter the name of the PF group for which you want to PCI Pass Through device.
        4. Interface Name - Displays the network interface name.
        5. Click Add to confirm.
      • For the PCI Device, add the following information.
        Note:
        1. Shared Across NUMA - Select the button to enable or disable sharing of the devices across NUMA.
          Note: Shared Across NUMA is applicable only when NUMA Alignments is enabled.
        2. Enter the name of the resource in Resource Name.
        3. Select the Target Driver from the drop-down menu.
          Note:

          Based on the Target driver, the system automatically adds the required Linux in Kernel Type, pciutils.

        4. PF Group - Enter the name of the PF group for which you want to PCI device.
        5. Add the number of PCI devices in Count.
        6. Click Add to confirm.

    • Kernel

      • Kernel Type - Select the Name and Version from the drop-down menu.
      • Kernel Arguments - Click Add to add a new kernel argument. Add the Key and Value in respective text box.
        Note: For hugepagesz and default_hugepagez, you can select the value from the drop-down menu. For other arguments, you can specify the key and value in respective text box.

      • Custom Packages - Click Add to add a new custom kernel package. Add the Name and Version in the respective text box.
    • Files - You can add a file or provide the file name as a source for file injection. Choose one of the following:
      • File - Select the file to be injected from the artifacts during CNF instantiation.
        To view the file in the drop-down menu, you must upload the file in the scripts folder. You can upload only .JSON, .XML, and .conf files.
        1. Click the Resources tab.
        2. Click the > icon corresponding to the root folder.
        3. Click the > icon corresponding to the Artifacts folder.
        4. Click the + icon corresponding to the scripts folder.
        5. Click Choose Files and select the file to upload.
        6. Click Upload to upload the selected file.
      • Input - Provide the <input name>, a meaningful name that serves as a reference to injecting the right file during CNF instantiation. For example, PTP4L_CONFIG_FILE, TS2PHC_CONFIG_FILE, and SYNCE4L_CONFIG_FILE.

        Once the user selects the file, it is injected into the specified location.

    Note: It is highly recommended to select File as the source.
    • Services - To add one or more of the following services, select the required service from the drop-down menu and click the add (+) icon.
      • stalld- A program also Known as 'stall daemon' is a mechanism to prevent the starvation of operating system threads in a Linux system.
      • syslog-ng - A log management solution that improves the performance of your SIEM solution by reducing the amount and improving the quality of data feeding your SIEM.

        When you select syslog-ng, the Add Service Config Files pop-up appears. Select the required configuration files for syslog-ng service.

      • phc2sys - A program that synchronizes two clocks in the system. It is used to synchronize the system clock to a PTP Hardware Clock (PHC). PHC is synchronized by the ptp4l(8) program.
      • ptp4l - An implementation of PTP according to IEEE standard 1588.
      • synce41 - A software implementation of Synchronous Ethernet (SyncE) according to ITU-T Recommendation G.8264.
      • ts2phc - Synchronizes PTP PHC to external time stamp signals. A single source may be used to distribute time to one or more PHC devices.
    • (Optional) Tuned Profiles - Enter the name of the tuned profile. You can add multiple tuned profiles separated by commas.
      Note: When you add a tuned profile, system adds the tuned package in the Custom Packages.
    • NUMA Alignments - Click the corresponding button to enable or disable the support for NUMA alignments.
    • Latency Sensitivity - You can set the latency value for high performance profiles. Select the value from the drop-down menu. You can select both high and low. Default value is normal.
    • I/O MMU Enabled - Click the corresponding button to enable or disable the I/O MMU.
    • Upgrade VM Hardware Version - Click the corresponding button to enable or disable upgrading the hardware version of the virtual machine.