Monitoring Points are lightweight software agents that execute the active synthetic probes necessary to monitor your network and web application performance. They are the source of all synthetic network traffic and can also be a target of testing.
Monitoring architecture
Cloud Network Insights uses a centralized management model with Monitoring Points.
- Google Cloud project - the central control where you enable Cloud Network Insights. You use this project to do the following:
- Download Monitoring Point installation packages (Docker, Helm, OVA/QCOW2, TF)
- View the status and inventory of all your deployed Monitoring Points
- Access Cloud Monitoring dashboards for metrics
- Configure Alert Policies and Notification Channels in Google Cloud Observability
- Monitoring locations - the target hosts in the environments you want to monitor. This is where you install the Monitoring Points.
Deployment locations
When installing Monitoring Points in the monitoring locations, select critical network segments, such as a central VPC, remote branch office, or specific cloud region. The following environments are supported:
- Google Cloud VPC networks in the same or different projects
- Remote branch offices or retail locations
- On-premises data centers
- Other cloud environments, for example, AWS or Azure
If you need help to determine where to install Monitoring Points to target
strategic web application deployment, use VPC Flow Logs or vm_flow metrics in
a performance dashboard to understand where traffic is flowing in your network.
We don't recommend installing Monitoring Points on the project with Cloud Network Insights enabled or directly on the workloads you want to monitor. For more information, see Best practices or AppNeta's Deployment Best Practices.
Supported Monitoring Point types
Cloud Network Insights supports the following Monitoring Point types.
| Monitoring Point type | Best for | Link to AppNeta documentation | |
|---|---|---|---|
| Containers | Docker container | Lightweight and ideal for cloud instances. Linux VMs, Google Compute Engine instances, or any Docker-enabled host, including on-premises environments. |
c50-Docker/Podman |
| Kubernetes (Helm) | Best for orchestrated environments. Google Kubernetes Engine (GKE), Amazon EKS, Azure AKS, or on-premises Kubernetes clusters. |
c50-Kubernetes (Helm) | |
| Virtual machines (VM) | VMware (OVA) | On-premises data centers running VMware ESXi or vCenter. Deploys as a full virtual machine. |
v35-VMware |
| KVM (QCOW2) | Monitor telco clouds or private Linux clouds. Linux-based virtualized environments such as OpenStack, custom Linux hypervisors. |
v35-KVM | |
| Cloud | Google Cloud | Seamlessly embed in your Google Cloud infrastructure. Deploys a Compute Engine VM using Terraform. |
Google CE Cloud Monitoring Point |
Before you begin
The deployment environment and Monitoring Points have specific networking requirements to ensure that the Monitoring Point can be installed successfully and communicate back to the Cloud Network Insights control plane.
Required roles
To get the permissions that you need to add Monitoring Points, ask your administrator to grant you the following IAM roles:
- Cloud Network Insights Editor (
roles/networkmanagement.CloudNetworkInsightsEditor) on the project that has Cloud Network Insights enabled -
Deploy Compute Engine Monitoring Points by using Terraform:
Compute Admin (
roles/compute.admin) on the project in which you're deploying the Compute Engine Monitoring Point -
Deploy a Monitoring Point by using Google Cloud Marketplace:
- Cloud Infrastructure Manager Admin (
roles/config.admin) on on the project in which you're deploying the Compute Engine Monitoring Point - Consumer Procurement Entitlement Manager (
roles/consumerprocurement.entitlementManager) on on the project in which you're deploying the Compute Engine Monitoring Point - Compute Admin (
roles/compute.admin) on on the deployment service account - Cloud Infrastructure Manager Agent (
roles/config.agent) on on the deployment service account - Service Account User (
roles/iam.serviceAccountUser) on on the deployment service account
- Cloud Infrastructure Manager Admin (
For more information about granting roles, see Manage access to projects, folders, and organizations.
You might also be able to get the required permissions through custom roles or other predefined roles.
If you want to use OS Login, specific IAM roles must be granted to the user accounts who connect to the VMs. See Assign OS Login IAM roles in the Compute Engine documentation.
Firewall requirements
Since we recommend installing Monitoring Points behind your firewalls, you might need to modify your firewall rules. Monitoring Points require outbound internet access to communicate back to the Cloud Network Insights control plane.
Protocol |
Port |
Description |
|---|---|---|
TCP |
443 (HTTPS) |
Required. Connectivity to the Cloud Network Insights control plane |
UDP |
123 (NTP) |
Required. Network Time Protocol. The Monitoring Point will fail to connect if time is not synced |
UDP/TCP |
53 (DNS) |
Required. Domain Name Service to resolve Cloud Network Insights endpoints |
UDP |
3239, 33434 |
Test Traffic. Required for standard Network Path monitoring (dual-ended) |
ICMP |
Type 8 (echo request) |
Test Traffic. Required for single-ended paths (for example, pinging 8.8.8.8). |
For more information, see Firewall Configuration from AppNeta.
Monitoring Point requirements
Review the AppNeta Monitoring Point specifications to ensure that you can successfully deploy them.
- Specifications – c50-Docker/Podman
- Specifications – c50-Kubernetes (Helm)
- Specifications – v35-VMware
- Specifications – v35-KVM
- Specifications – Google CE Cloud Monitoring Point
Download installation package and deploy
When you're ready to deploy Monitoring Points in your network, do the following:
- Download the Monitoring Point installation bundle from Google Cloud.
- Install the Monitoring Point.
Deploy containers (c50)
Get the installation bundles, and then install them on the target host.
Download Monitoring Point installation bundle
Get the installation bundle and ensure that the bundle is on the target host.
Console
If you have administrator credentials to the target host or if you need to download the installation bundle locally before transferring the file to the target host, use the Google Cloud console.
In the Google Cloud console, navigate to Network Intelligence > Cloud Network Insights > Monitoring Points.
Click Add monitoring point.
In the Platform Type list, select Docker/Podman or Helm (Kubernetes), and then click Continue.
Enter the Hostname of the container. This is the name of the host you're installing the Monitoring Point on.
In Download the Monitoring Point, choose how you want to download the installation bundle.
- To download the installation bundle on the target host using a terminal,
select Run a command on the host.
- Click Open shell and generate token. You might need to authorize Cloud Shell.
- Press Enter to generate a token, and then copy and paste it in the text field.
- Open a terminal to your target host using administrator credentials, and run the generated command to download the installation bundle. The curl command is valid for one hour.
- To download the installation bundle locally, select Download locally.
- Click Download Monitoring Point.
- Move the installation bundle to the target host.
- To download the installation bundle on the target host using a terminal,
select Run a command on the host.
gcloud
Use this method if your target host has Google Cloud CLI installed.
gcloud network-management network-monitoring-providers \ monitoring-points download-install-script \ --network-monitoring-provider=PROVIDER_NAME --location=global \ --monitoring-point-type=MP_TYPE \ --hostname=HOST_NAME \ --output-file=compose.HOST_NAME.tar.gz
Replace the following:
PROVIDER_NAME: the provider name,externalby default.MP_TYPE: the Monitoring Point container type. Use one of the following values:- Enter
CONTAINERfor Docker or Podman - Enter
HELMfor Helm (Kubernetes)
- Enter
HOST_NAME: the name of the host you're installing the Monitoring Point on.
Install your container-based Monitoring Point
Once the install bundle is on the target host, follow the instruction from AppNeta to install the Monitoring Point.
You can't change the configuration settings for containers. If you need to change the proxy, NTP, or timezone settings later, you delete the Monitoring Point and add a new one.
Deploy VMs (v35)
Get the Monitoring Point images and then deploy them using VMware vSphere (vCenter or ESXi) or Linux KVM (QCOW2).
Download Monitoring Point image and configuration
You can download a zip file locally with the Monitoring Point images and configuration files or generate a command to download the installation bundle on the machine you're installing the Monitoring Point on.
In the Google Cloud console, navigate to Network Intelligence > Cloud Network Insights > Monitoring Points.
Click Add monitoring point.
In the Platform Type list, select either KVM or VMware, and then click Continue.
Enter the Password for the VM.
Choose your IP Configuration.
- If the target network assigns IP addresses automatically, select Obtain settings using DHCP.
- If you want to use a static IP address, select Static IP.
Complete the following IP configuration fields. Fields are based on what you selected for the IP Configuration.
- IP Address: The IPv4 address that you want to assign to the Monitoring Point.
- Network Mask:The subnet mask (for example, 255.255.255.0).
- Network Gateway: The default gateway IP address.
- DNS 1: The IPv4 address of your primary DNS servers (for example, 8.8.8.8).
- DNS 2 (optional): A backup name server to ensure connectivity if the primary fails (for example, 1.1.1.1)
- Domain name: The default DNS search domain for the Monitoring Point.
- Hostname: The unique name for this Monitoring Point.
- Timezone (optional): The timezone of the VM the Monitoring Point is installed on.
- NTP 1 (optional): The primary Network Time Protocol address (IP or FQDN).
- NTP 2 (optional): A backup time server used if the primary address is unreachable.
Click Continue.
In Download the Monitoring Point, choose how you want to download the installation bundle.
- To download the installation bundle locally, select Download locally.
- Click Download Monitoring Point.
- For KVM: Download a ZIP file that contains an ISO file (boot/config drive).
- For VMware: Download a ZIP file that contains a JSON file (deployment parameters).
- Download the disk image. The available options depends on your selected Platform Type.
- Move the configuration file and the disk image to your hypervisor management station.
- Click Download Monitoring Point.
- To generate a command-line command to run on the host, select
Run a command on the host.
- Click Open shell and generate token. You might need to authorize Cloud Shell.
- Press Enter to generate a token, and then copy and paste it in the text field in Generate the command.
- Open a terminal to your host using administrator credentials, and run the generated command to download the configuration files.
- Download the disk image. The available options depends on your selected Platform Type.
- If necessary, move the configuration file and the disk image to your hypervisor management station.
- To download the installation bundle locally, select Download locally.
Install VM-based Monitoring Point
Once you have the images, follow the instructions from AppNeta to deploy the Monitoring Point.
After deploying the VM, record the MAC Address of the primary network adapter (eth0 or Network Adapter 1) from your hypervisor (vCenter or KVM).
- Username: admin
- Default Password: The MAC address in UPPERCASE with no colons (for example, 00505681450A).
You need these credentials if you need to sign in to the Monitoring Point's Web Administration interface to troubleshoot connectivity issues. For more information, see AppNeta's documentation.
Configure a web proxy
If your network requires a proxy for outbound access, you must configure it after the VM is powered on.
Deploy Compute Engine VMs
You can deploy a Compute Engine VM to Google Cloud by using Terraform. The Terraform configuration files (TF) are available through the Google Cloud console or through Google Cloud Marketplace.
Console
In the Google Cloud console, navigate to Network Intelligence > Cloud Network Insights > Monitoring Points.
Click Add monitoring point.
In the Platform Type list, select Google Cloud, and then click Continue.
Enter the Hostname of the VM. This name is used as the VM name and as the Monitoring Point's name by default.
Optional: Select Connect via Private Service Connect (PSC) so that the Monitoring Point can communicate with AppNeta using Private Service Connect. You must already have Private Service Connect endpoints in your Virtual Private Cloud.
For more information, see the AppNeta documentation.
Click Continue.
In Select a Monitoring Point deployment option, select Terraform.
In Download the Monitoring Point, choose how you want to download the installation bundle.
- To download the installation bundle locally, select Download locally.
- Click Download Monitoring Point.
- Extract the archive in the directory where you want to deploy the Terraform configuration files.
- To download the installation bundle on a host using a terminal,
select Run a command on the host.
- Click Open shell and generate token. You might need to authorize Cloud Shell.
- Press Enter to generate a token, and then copy and paste it in the text field.
- Open a terminal to your host using administrator credentials, and run the generated command to download an archive of the Terraform configuration files. The curl command is valid for one hour.
- To download the installation bundle locally, select Download locally.
Marketplace
In the Google Cloud console, navigate to Network Intelligence > Cloud Network Insights > Monitoring Points.
Click Add monitoring point.
In the Platform Type list, select Google Cloud, and then click Continue.
Enter the Hostname of the VM. This name is used as the VM name and as the Monitoring Point's name by default.
Optional: Select Connect via Private Service Connect (PSC) so that the Monitoring Point can communicate with AppNeta using Private Service Connect. You must already have Private Service Connect endpoints in your Virtual Private Cloud.
For more information, see the AppNeta documentation.
Click Continue.
Select Google Cloud Marketplace, and select the project you want to deploy the Monitoring Point in.
Copy the configuration code snippet for the Monitoring Point. If you've deployed a Monitoring Point from Cloud Marketplace before, the code snippet is entered in the Configuration field automatically.
Click Open through Google Cloud Marketplace.
You might need to review the Marketplace Terms of Service. To install Cloud Network Insights using Google Cloud Marketplace accept the Terms of service.
Click either the Terraform tab or the Command-Line Deployment tab and follow the on-screen instructions.
- In Network interfaces, configure an external IP address to allow the Monitoring Point to communicate with AppNeta.
- In Configuration, paste the Monitoring Point configuration code snippet.
Click Deploy.
Install with Terraform
Once the install bundle is on the host, follow the instructions from AppNeta to install the Monitoring Point.
For more information about deploying a VM from Google Cloud Marketplace, see either Deploy a software package or Deploy virtual machine (VM) products using the command-line interface (CLI).
Verify the install
Open the Google Cloud console and navigate to Network Intelligence Center > Cloud Network Insights. After 2–5 minutes, the Monitoring Point appears in the table with the status Active. If it doesn't appear within ten minutes, see Troubleshooting.
Remove Monitoring Points
To remove a Monitoring Point, delete it in AppNeta. For more information about removing a Monitoring Point, see the AppNeta documentation.
What's next
After you add Monitoring Points, it's time to create monitoring policies to populate Cloud Network Insights with data.