Intro

In my previous [post]({{< ref "post/12-opnsense-virtualization-highly-available" >}}), I've set up a PoC to validate the possibility to create a cluster of 2 OPNsense VMs in Proxmox VE and make the firewall highly available.

This time, I will cover the creation of my future OPNsense cluster from scratch, plan the cut over and finally migrate from my current physical box.

Build the Foundation

For the real thing, I'll have to connect the WAN, coming from my ISP box, to my main switch. For that I have to add a VLAN to transport this flow to my Proxmox nodes.

UniFi

The first thing I do is to configure my layer 2 network which is managed by UniFi. There I need to create two VLANs:

• WAN (20): transport the WAN between my ISP box and my Proxmox nodes.
• pfSync (44), communication between my OPNsense nodes.

In the UniFi controller, in Settings > Networks, I add a New Virtual Network. I name it WAN and give it the VLAN ID 20:

I do the same thing again for the pfSync VLAN with the VLAN ID 44.

I will plug my ISP box on the port 15 of my switch, which is disabled for now. I set it as active, set the native VLAN on the newly created one WAN (20) and disable trunking:

Once this setting applied, I make sure that only the ports where are connected my Proxmox nodes propagate these VLAN on their trunk.

We are done with UniFi configuration.

Proxmox SDN

Now that the VLAN can reach my nodes, I want to handle it in the Proxmox SDN.

In Datacenter > SDN > VNets, I create a new VNet, name it vlan20 to follow my own naming convention, give it the WAN alias and use the tag (ID) 20:

I also create the vlan44 for the pfSync VLAN, then I apply this configuration and we are done with the SDN.

Create the VMs

Now that the VLAN configuration is done, I can start buiding the virtual machines on Proxmox.

The first VM is named cerbere-head1 (I didn't tell you? My current firewall is named cerbere, it makes even more sense now!) Here are the settings:

• OS type: Linux
• Machine type: q35
• BIOS: OVMF (UEFI)
• Disk: 20 GiB on Ceph storage
• CPU/RAM: 2 vCPU, 4 GiB RAM
• NICs:
  1. vmbr0 (Mgmt)
  2. vlan20 (WAN)
  3. vlan13 (User)
  4. vlan37 (IoT)
  5. vlan44 (pfSync)
  6. vlan55 (DMZ)
  7. vlan66 (Lab)

ℹ️ Now I clone that VM to create cerbere-head2, then I proceed with OPNsense installation. I don't want to go into much details about OPNsense installation, I already documented it in the previous [post]({{< ref "post/12-opnsense-virtualization-highly-available" >}}).

After the installation of both OPNsense instances, I give to each of them their IP in the Mgmt network:

• cerbere-head1: 192.168.88.2/24
• cerbere-head2: 192.168.88.3/24

While these routers are not managing the networks, I give them my current OPNsense routeur as gateway (192.168.88.1) to able to reach them from my PC in another VLAN.

Configure OPNsense

Initially I thought about restoring my current OPNsense config on the VM. But as I didn't document the configuration process the first time, I take the opportunity to start over.

I'll start with the elements that needs to be configured on both firewalls, where each has its own parameters. After I'll create the OPNsense cluster, then configure the master node only as the configuration will be duplicated on the other node.

System

I start by the basic, in System > Settings > General:

• Hostname: cerbere-head1 (cerbere-head2 for the second one).
• Domain: mgmt.vezpi.com.
• Time zone: Europe/Paris.
• Language: English.
• Theme: opnsense-dark.
• Prefer IPv4 over IPv6: tick the box to prefer IPv4.

Then, in System > Access > Users, I create a new user, I don't like sticking with the defaults root. I add this user in the admins group, while removing root from it.

In System > Settings > Administration, I change several things:

• TCP port: from 443 to 4443, to free port 443 for the reverse proxy coming next.
• Alternate Hostnames: cerbere.vezpi.com which will be the URL to reach the firewall by the reverse proxy.
• Access log: enabled.
• Secure Shell Server: enabled.
• Authentication Method: permit password login (no root login).
• Sudo: No password. Once I click Save, I follow the link given to reach the WebGUI on port 4443.

Time for updates, in System > Firmware > Status, I click on Check for updates. An update is available, I close the banner, head to the bottom and click on Update. I'm warned that this update requires a reboot.

Once updated and rebooted, I go to System > Firmware > Plugins, I tick the box to show community plugins. For now I only install the QEMU guest agent, os-qemu-guest-agent, to allow communication between the VM and the Proxmox host.

This requires a shutdown. On Proxmox, I enable the QEMU Guest Agent in the VM options:

Finally I restart the VM. Once started, from the Proxmox WebGUI, I can see the IPs of the VM which confirms the guest agent is working.

Interfaces

On both firewalls, I assign the remaining NICs to new interfaces adding a description. The VMs have 7 interfaces, I carefully compare the MAC addresses to not mix them up:

In the end, the interfaces configuration looks like this:

High Availability

From here we can associate both instances to create a cluster. The last thing I need to do, is to allow the communication on the pfSync interface. By default, no communication is allowed on the new interfaces.

From Firewall > Rules > pfSync, I create a new rule on each firewall:

• Action: Pass
• Quick: tick the box to apply immediately on match
• Interface: pfSync
• Direction: in
• TCP/IP Version: IPv4
• Protocol: any
• Source: pfSync net
• Destination: pfSync net
• Log: tick the box to log packets
• Category: OPNsense
• Description: pfSync

Next, I head to System > High Availability > Settings:

• Master (cerbere-head1):
  • Synchronize all states via: pfSync
  • Synchronize Peer IP: 192.168.44.2
  • Synchronize Config: 192.168.44.2
  • Remote System Username: <username>
  • Remote System Password: <password>
  • Services: Select All
• Backup (cerbere-head2):
  • Synchronize all states via: pfSync
  • Synchronize Peer IP: 192.168.44.1
  • Synchronize Config: 192.168.44.1 ⚠️ Do not fill the XMLRPC Sync fields, only to be filled on the master.

In the section System > High Availability > Status, I can verify is the synchronization is working. On this page I can replicate any or all services from my master to my backup node:

Virtual IPs

Now that HA is configured, I can give my networks a VIP shared across my nodes. In Interfaces > Virtual IPs > Settings, I create one VIP for each of my networks using CARP (Common Address Redundancy Protocol). The target is to reuse the IP addresses used by my current OPNsense instance, but as it is still routing my network, I use different IPs for the configuration phase:

Firewall

Let's configure the core feature of OPNsense, the firewall. I don't want to go too crazy with the rules. I only need to configure the master, thanks to the replication.

Basically I have 2 kinds of networks, those which I trust, and those which I don't. From this standpoint, I will create two zones.

Globally, on my untrusted networks, I will allow access to the DNS and to the internet, not on the other networks. On the other hand, my trusted networks would have the possibility to reach other VLANs.

To begin, in Firewall > Groups, I create 2 groups to regroup my interfaces:

• Trusted: Mgmt, User
• Untrusted: IoT, DMZ, Lab

Next, in Firewall > Aliases, I create an alias InternalNetworks to regroup all my internal networks:

For all my networks, I want to allow DNS querry on the local DNS. In Firewall > Rules > Floating, let's create the first rule:

Next I want to allow connections towards the internet. At the same place I create a second rule:

Finally, I want to allow anything from my trusted networks. In Firewall > Rules > Trusted, I create the rule:

Great, with these 3 rules, I cover the basics. The remaining rules would be to allow specific equipment to reach out to something else. For example my home assistant instance want to connect to my TV, both are on different VLAN, hence I need a rule to allow it. I won't cover that in this post.

DHCP

For the DHCP, I choose Dnsmasq. In my current installation I use ISC DHCPv4, but as it is now deprecated, I prefer to replace it. Dnsmasq will also act as DNS, but only for my local zones.

Beware because it is not synchronizing leases in HA. To workaround this, both firewalls will serve DHCP at the same time, with slight different configuration to not overlap.

In Services > Dnsmasq DNS & DHCP > General, I configure the master firewall as follow:

• Default
  • Enable: Yes
  • Interface: Mgmt, User, IoT, DMZ and Lab
• DNS
  • Listen port: 53053
• DNS Query Forwarding
  • Do not forward to system defined DNS servers: Enabled
• DHCP
  • DHCP FQDN: Enabled
  • DHCP local domain: Enabled
  • DHCP authoritative: Enabled
  • DHCP reply delay: 0
  • DHCP register firewall rules: Enabled
  • Disable HA sync: Enabled

On the backup node, I configure it the same, the only difference will be the DHCP reply delay which I set to 10. This will let the time to my master node to fulfill requests if it is alive.

Next I configure the DHCP ranges. Both firewalls will have different ranges, the backup node will have smaller ones. On the master, they are configured as follow:

Then I set some DHCP options for each domain: the router, the dns-server and the domain-name:

Finally in in the Hosts tab, I define static DHCP mappings but also static IP not managed by the DHCP, to have them registered in the DNS:

DNS

For the DNS, I will use Unbound. It is a validating, recursive, caching DNS resolver built into OPNsense, which can:

• Resolve queries from the root servers.
• Cache results for faster responses.
• Check domain authenticity with DNSSEC.
• Block domains based of blacklist.
• Add custom records.

For the local zones, I will use forward the requests to Dnsmasq, hence I will not registering DHCP leases in Unbound.

Let's configure it, in Services > Unbound DNS > General:

Then I configure the blocklist in Services > Unbound DNS > Blocklist. I enable it and select the [hagezi] Multi PRO mini list. Initially I was using AdGuard Home, but I want to give this blocklist feature a chance.

Finally I configure query forwarding for my local domains. In Services > Unbound DNS > Query Forwarding, I add each of my local domains with their reverse lookup (PTR record). The forwarded server is Dnsmasq which I'll configure next:

VPN

When I'm not home, I still want to be able to reach my services and enjoy my DNS ad blocker. For that I'm setting up a VPN, with WireGuard. It's fast, secure and easy to set up.

In VPN > WireGuard > Instances, I create a new one:

• Enabled: Yes
• Name: Homelan
• Public/Private keys: Key-pair generated
• Listen port: 61337
• Tunnel address: 10.13.37.1/24
• Depend on (CARP): on lan (vhid 1)

Once configured, I enable WireGuard and apply the configuration.

Next in the Peer generator tab, I fulfill the empty fields for my first device:

• Endpoint: vezpi.com
• Name: S25Ultra
• DNS Servers: 10.13.37.1

Before clicking Store and generate next, from my device, I configure the peer by capturing the QR code. Finally I can save that peer and start over for new ones.

Reverse Proxy

The next feature I need is a reverse proxy, to redirect incoming requests, to reach my services, such as this blog. For that I use Caddy. This service is not installed by default, I need to add a plugin.

On both firewalls, In System > Firmware > Plugins, I tick the box to show community plugins and install os-caddy.

I refresh the page and, on the master, in Services > Caddy > General Settings:

• Enable Caddy: Yes
• Enable Layer4 Proxy: Yes
• ACME: <my mail address>
• Auto HTTPS: On (default)

mDNS Repeater

Switch

Enable DHCP Change VIP Replicate configuration

Verify

Firewall All sites mDNS (chromecast) VPN

DNS blocklist