Preparing for deployment

Information to gather before deployment, choosing a deployment method, and pre-deployment steps shared by all deployment methods.

Preparing to Deploy Nutanix vSensors

Basic information to collect:

• IP address and subnet mask for the Management interface of the vSensor.

  • DHCP is also supported.
• IP address or hostname of the Vectra Brain that you will be pairing with.

  • The IP address or hostname of the Brain will be baked into the vSensor image downloaded from your Vectra UI and will point to the Brain associated with your Vectra UI.

    • When the Sensor powers on, it will reach out to Brain using this selection to announce itself as Available for pairing.
  • Whether this points to the IP of the Brain or the hostname depends on what you have set in your Brain under Configuration → COVERAGE → Data Sources → Network → Sensors → Sensor Configuration → Sensor Registration and Pairing for Pair using the Detect Brain.

Please Note:

You must have a DNS Name (host name) is configured in Configuration → COVERAGE → Data Sources → Network → Brain Setup → Brain to see the DNS Name option in Sensor Pairing and Registration.

It is recommended to use hostname for pairing when possible. Please see pairing Nutanix vSensors for more details.

• DNS server addresses.
• To configure your vSensor, you will need access to the vSensor Command Line Interface (CLI) either via the console in your hypervisor or via SSH.

  • DHCP is enabled by default upon vSensor initial boot.
  • You must know the IP that was assigned via DHCP to SSH to the CLI, otherwise you will need to use the hypervisor console to access the CLI and perform initial configuration.
• For production monitoring, ensure that the vSensor VM is kept running 24x7, and ensure that the hypervisor does not overcommit resources or otherwise misrepresent the resources it is providing to the vSensor.

Nutanix specific guidance:

• You must have access to both Prism Central and Prism Element for Service Chaining 1.0

  • You must have permissions to make Prism Central REST API calls for Service Chaining 1.0.
• Traffic Mirroring only requires the Prism Central UI for deployment.

  • Setting the vSensor as an Agent VM will require Prism Element or ACLI commands.
• You must have access to a Controller VM (CVM) ACLI command line to perform ACLI commands.

Choosing a Deployment Method

• Both deployment methods will share the pre-deployment steps below.
• You can use both deployment methods if you desire when deploying multiple Nutanix vSensors.

  • For example:

    • You might use one vSensor deployed using Traffic Mirroring to capture physical network traffic that arrives on a host port or host bonded port on a particular node.
    • You might use another vSensor using Service Chaining 1.0 to capture VM to VM traffic and all other subnet traffic.
  • More than one vSensor can be deployed on a single node in your Nutanix cluster if required.

Links for either deployment method:

• Service Chaining 1.0 deployment
• Traffic Mirroring deployment
• For either deployment method, ensure you follow the recommend post deployment configuration steps.

Pre-Deployment Steps for the Nutanix vSensor

Downloading the Nutanix vSensor Image

The Nutanix vSensor image is available under Configuration → COVERAGE → Data Sources → Network → Sensors in your Vectra UI. Navigate to this area and then click Download Virtual Image at the top right and select the Nutanix vSensor (QCOW2) option.

Extract vSensor Image Files

The downloaded vSensor image will have a filename similar to this (updates to the image after this documentation was written will change this):

• vectra-vsensor-117678-gee9a02c9f7-1_9.0.0-18-57-192_168_52_73-default.tar.gz

Using the unarchiving tool of your choice, decompress the .gz file. Then untar the file to a directory of your choice. That directory will contain 3 files such as:

The vectra-vsensor.sh file can be ignored for Nutanix vSensor deployment. The base image is shared for both Nutanix and KVM deployment. KVM deployment uses that file, but Nutanix only needs the .iso and .qcow2 files.

Upload vSensor Image Files to Nutanix

If using Prism Element for the initial VM deployment (Service Chaining 1.0 Only):

Use the image service available in Prism Element at Settings → Image Configuration → + Upload Image to upload the seed.iso and vectra-vsensor.qcow2 images. You can name these how you prefer in Nutanix.

• Please note that the seed.iso should be set to an Image Type of ISO and the .qcow2 should have an Image type of Disk.
• Once the images have become Active, you can proceed to create the VM.

Please Note:

The .qcow2 image will take considerably longer to become active than the seed.iso.

If using Prism Central for the initial VM deployment (for Service Chaining 1.0 and Traffic Mirroring):

Navigate to Infrastructure → Compute → Images → Add Image, and upload the seed.iso and vectra-vsensor.qcow2 images.

The Image Source should always be Image File and when you select the seed.iso or .qcow disk file they will automatically be recognized as ISO or DISK Image types.

Unlike when using Prism Element to upload the images, you do NOT need to wait for the images to become Active, but you do need to wait for them to finish uploading. When you see the file sizes appear in the image library, you know they are ready to deploy:

Examples of the upload dialogs for Prism Central: