Skip to content

Terraform tutorial detailing how to deploy Active/Passive VM-Series with session sync in Google Cloud.

License

Notifications You must be signed in to change notification settings

PaloAltoNetworks/google-cloud-vmseries-ha-tutorial

Repository files navigation

VM-Series Active/Passive HA on Google Cloud

This tutorial creates a pair of Active/Passive VM-Series firewalls on Google Cloud. This deployment model provides solutions for the following key use-cases:

  • Maintaining session continuity through stateful failover between the VM-Series firewalls.
  • Terminating IPsec tunnels directly to the VM-Series firewall through the external load balancer.
  • Preserving the original client IP address for internet inbound traffic to internal applications protected by the VM-Series firewalls.

The autoscale architecture is recommended in most use-cases. Please see VM-Series on Google Cloud for more information on VM-Series deployment models.

Architecture

In this model, each VM-Series firewall belongs to an unmanaged instance group. Only the primary VM-Series firewall receives network traffic from Google Cloud load balancers. The health check configured on the load balancers determines the HA state of the primary VM-Series firewall. If the health check fails on the primary VM-Series firewall, the load balancers carry the active sessions to the secondary VM-Series firewall. At that point, the secondary VM-Series firewall becomes the primary firewall.

Note

During a failure event, sessions are carried between the firewalls via connection tracking on the load balancers.

Overview Diagram

Resources
VM-Series
  • 2 x VM-Series are deployed to separate zones within a region.
  • Each firewall belongs to an unmanaged instance group.
  • Each instance group is a backend service of an internal load balancer.
External LB An external pass-through load balancer with two forwarding rules:
  • Rule 1: Forwards internet inbound traffic to the VM-Series untrust interface.
  • Rule 2: Forwards outbound traffic from the untrust interfaces to the internet.
Internal LB An internal pass-through load balancer with a single forwarding rule. Traffic from workload networks use the load balancer's forwarding rule as the next hop within their VPC route table.
VPC Networks
Mgmt VPC Contains the VM-Series MGT interfaces. This interface also serves as the HA1 interface.
HA2 VPC Contains the VM-Series HA2 interfaces.
Untrust VPC Serves as the internet gateway for resources within the trust VPC.
Trust VPC Contains the workloads protected by the VM-Series. This VPC can also serve as a hub network with multiple VPCs peered to it.

Important

After deployment, the load balancer's health checks will only pass on the active VM-Series firewall. This is because the dataplane of the passive firewall is inactive and is unable to pass the health checks. During a failure event, the passive firewall becomes active and the health checks will pass.

Prepare for deployment

  1. Enable the required APIs, generate an SSH key, and clone the repository.

    gcloud services enable compute.googleapis.com
    ssh-keygen -f ~/.ssh/vmseries-tutorial -t rsa
    git clone https://github.com/PaloAltoNetworks/google-cloud-vmseries-ha-tutorial
    cd google-cloud-vmseries-ha-tutorial
    
  2. Create a terraform.tfvars file.

    cp terraform.tfvars.example terraform.tfvars
    
  3. Edit the terraform.tfvars file and set values for the following variables:

    Variable Description
    project_id Set to your Google Cloud deployment project.
    public_key_path Set to match the full path you created previously.
    mgmt_allow_ips Set to a list of IPv4 ranges that can access the VM-Series management interface.
    prefix (Optional) If set, this string will be prepended to the created resources.
    vmseries_image_name (Optional) Defines the VM-Series image to deploy. A full list of images can be found here.
  4. (Optional) If you are using BYOL image (i.e. vmseries-flex-byol-*), the license can be applied during deployment by adding your VM-Series authcode to bootstrap_files/authcodes.

  5. Save your terraform.tfvars file.

Deployment

When no further changes are necessary in the configuration, deploy the resources:

  1. Initialize and apply the Terraform plan.

    terraform init
    terraform apply
    
  2. Enter yes to start the deployment.

  3. After all the resources are created, Terraform displays the following message:

    Apply complete!
    
    Outputs:
    
    EXTERNAL_LB_IP     = "ssh paloalto@1.1.1.1 -i ~/.ssh/vmseries-tutorial"
    EXTERNAL_LB_URL    = "https://1.1.1.1"
    VMSERIES_ACTIVE    = "https://2.2.2.2"
    VMSERIES_PASSIVE   = "https://3.3.3.3"
    VMSERIES_SSH       = "ssh admin@1.1.1.1 -i ~/.ssh/vmseries-tutorial" 
    

Access the Active VM-Series Firewall

To access the VM-Series user interface, a password must be set for the admin user.

Caution

After the apply completes, it may ~10 minutes for the VM-Series to become available.

  1. SSH to the active VM-Series firewall using the VMSERIES_SSH output value.

    ssh admin@1.1.1.1 -i ~/.ssh/vmseries-tutorial"
    
  2. On the active VM-Series, set a password for the admin username.

    configure
    set mgt-config users admin password
    
  3. Commit the changes.

    commit
    
  4. Enter exit twice to terminate the session.

Test the Deployment

You can now test the deployment by accessing the workload-vm that resides in the trust VPC network. All of the workload-vm traffic is routed directly through the VM-Series HA pair.

  1. Use the output EXTERNAL_LB_URL to access the web service on the workload-vm through the VM-Series firewall.

  2. Use the output EXTERNAL_LB_SSH to open an SSH session through the VM-Series to the workload-vm.

    ssh paloalto@1.1.1.1 -i ~/.ssh/vmseries-tutorial
    

Note

The address within EXTERNAL_LB_URL & EXTERNAL_LB_SSH is the inbound forwarding rule on the external load balancer. The VM-Series inspects and translates the request from the inbound forwarding rule to the workload-vm.

  1. On the workload VM, run a preloaded script to test the failover mechanism across the VM-Series firewalls.

    /network-check.sh
    

    You will see output like this where x.x.x.x is the IP address is EXTERNAL_LB_IP address.

    Wed Mar 12 16:40:18 UTC 2023 -- Online -- Source IP = x.x.x.x
    Wed Mar 12 16:40:19 UTC 2023 -- Online -- Source IP = x.x.x.x
    Wed Mar 12 16:40:20 UTC 2023 -- Online -- Source IP = x.x.x.x
    Wed Mar 12 16:40:21 UTC 2023 -- Online -- Source IP = x.x.x.x
    

Note

Egress traffic from the workload-vm is routed to the internal load balancer's forwarding rule. The VM-Series inspects and translates the request to the outbound forwarding rule on the external load balancer.

  1. Login to the VM-Series firewalls using the VMSERIES_ACTIVE and VMSERIES_PASSIVE output values.

  2. After login, take note of the HA Status in the bottom right corner on each firewall.

    Active Firewall

    Passive Firewall

  3. Perform a user initiated failover.

    1. On the Active Firewall, go to the Device → High Availability → Operational Commands.
    2. Click Suspend local device for high availability.
    3. When prompted, click OK to initiate the failover.
  4. You should notice your SSH session to the workload-vm is still active. This indicates the session successfully failed over between the VM-Series firewalls.

    Wed Mar 12 16:47:18 UTC 2023 -- Online -- Source IP = x.x.x.x
    Wed Mar 12 16:47:19 UTC 2023 -- Online -- Source IP = x.x.x.x
    Wed Mar 12 16:47:21 UTC 2023 -- Offline
    Wed Mar 12 16:47:22 UTC 2023 -- Offline
    Wed Mar 12 16:47:23 UTC 2023 -- Online -- Source IP = x.x.x.x
    Wed Mar 12 16:47:24 UTC 2023 -- Online -- Source IP = x.x.x.x
    

(Optional) Onboard Internet Applications

You can secure multiple internet facing applications through the VM-Series firewall. This is done by mapping the addreses/ports from external forwarding rules to NAT policies defined on the VM-Series firewall.

In this section, onboard a new web application by creating a forwarding rule on the external load balancer along with a corresponding NAT policy on the VM-Series.

Create an application VM

In Cloud Shell, deploy new virtual machine (app-vm) to a subnet within the trust VPC network. The app-vm installs a simple web application through its metadata startup script (/scripts/app_startup.sh).

  1. In Cloud Shell, set environment variables for your PROJECT_ID, REGION, ZONE, & SUBNET for the application environment.

     export PROJECT_ID=YOUR_PROJECT_ID
     export REGION=YOUR_REGION
     export ZONE=YOUR_ZONE
     export SUBNET=SUBNET_NAME

Tip

The app-vm can be deployed to any subnet in the trust-vpc, or to any VPC connected & routed to the trust-vpc (i.e. VPC peering).

  1. Create the app-vm virtual machine.

    gcloud compute instances create app-vm \
        --project=$PROJECT_ID \
        --network-interface subnet=$SUBNET,no-address \
        --zone=$ZONE \
        --machine-type=n2-standard-2 \
        --image-project=debian-cloud \
        --image-family=debian-11 \
        --metadata startup-script-url=https://raw.githubusercontent.com/PaloAltoNetworks/google-cloud-vmseries-ha-tutorial/main/scripts/app_startup.sh
    
  2. Record the INTERNAL_IP address of the new virtual machine.

    Output

     NAME: app-vm
     ZONE: us-central1-a
     MACHINE_TYPE: f1-micro
     PREEMPTIBLE:
     INTERNAL_IP: 10.0.2.4
     EXTERNAL_IP:
     STATUS: RUNNING

Note

In the VM-Series NAT policy, the INTERNAL_IP will be set as the translated packet's destination address.

Create Forwarding Rule

Create a forwarding rule (fwd-rule-app-vm) on the external load balancer. This rule will be used to distribute internet inbound traffic destined to the app-vm through the VM-Series untrust interfaces.

  1. Assign the name of the external load balancer's backend service to an environment variable named EXTERNAL_LB.

    export EXTERNAL_LB=$(gcloud compute backend-services list \
        --filter="loadBalancingScheme:EXTERNAL" \
        --format="get(name)")
    
    echo $EXTERNAL_LB
    
  2. Create a new forwarding rule on the external load balancer.

    gcloud compute forwarding-rules create fwd-rule-app-vm \
        --load-balancing-scheme=EXTERNAL \
        --region=$REGION \
        --ip-protocol=L3_DEFAULT \
        --ports=ALL \
        --backend-service=$EXTERNAL_LB
    
  3. Retrieve the forwarding rule's address.

    gcloud compute forwarding-rules describe fwd-rule-app-vm \
        --region=$REGION \
        --format='get(IPAddress)'
    

Note

In the firewall's NAT rule, the forwarding rule address is set as the original packet's destination address.

Create NAT Policy

On the VM-Series, create a NAT policy to translate traffic destined the forwarding rule (fwd-rule-app-vm) to the internal IPv4 address of the app-vm IP address (i.e. 10.0.2.4).

  1. On the active VM-Series, go to Policies → NAT.

  2. Click Add and enter a name for the rule (i.e. inbound-app-vm).

  3. Configure the Original Packet as follows:

    Field Value
    Source Zone untrust
    Destination Zone untrust
    Service service-http
    Destination Address The forwarding rule's IPv4 adddress (i.e. 34.172.143.223).
  4. In the Translated Packet tab, configure the Destination Address Translation as follows:

    Field Value
    Translation Type Static IP or Dynamic IP
    Translated Address The INTERNAL_IP of the app-vm (i.e. 10.0.2.4)
  5. Click OK to create the rule.

  6. Click Commit to apply the changes.

  7. After the commit completes, access the sample application using the address of the forwarding rule.

    http://34.172.143.223
    
  8. On the active VM-Series, go to Monitor → Traffic to view the traffic destined to the app-vm.

Clean up

To avoid incurring charges to your Google Cloud account for the resources you created in this tutorial, delete all the resources when you no longer need them.

  1. (Optional) If you did the Onboard Internet Applications section, delete the forwarding rule (fwd-rule-app-vm) and application VM (app-vm).

    gcloud compute forwarding-rules delete fwd-rule-app-vm \
        --region=$REGION \
        --quiet
    
    gcloud compute instances delete app-vm \
        --zone=$ZONE \
        --quiet
    
  2. Run the following command to delete the resources.

    terraform destroy
    

    Enter yes to delete the resources.

  3. After all the resources are deleted, Terraform displays the following message.

    Destroy complete!
    

Additional information