Configure a failover IP

A failover IP is an address routed to your dedicated server, movable to another machine on failure or migration. HolyCloud associates it in the panel; you configure the network interface under Linux.

Prerequisites

• HolyCloud dedicated server with working main IP
• Failover IP ordered and visible in the client area
• root SSH access
• Interface name (ip link) — often eno1, eth0, enp3s0f0

Step 1: associate the IP in the panel

1. Client area → Dedicated servers → your machine.
2. IP / Failover section.
3. Attach the failover IP to this server (detach from another server if it was there).
4. Note: IP address, netmask (often /32 or /255.255.255.255), gateway if shown.

HolyCloud failover routing is usually via ARP or routing to your MAC — follow panel docs for your infrastructure generation.

Step 2: temporary configuration (test)

ip addr show
sudo ip addr add IP_FAILOVER/32 dev eth0
ping -c 3 IP_FAILOVER

Replace eth0 with your public interface.

If ping OK from an external host (after firewall rules), move to persistent config.

Step 3: Debian/Ubuntu (ifupdown)

File /etc/network/interfaces.d/failover.cfg:

auto eth0:0
iface eth0:0 inet static
    address IP_FAILOVER
    netmask 255.255.255.255

Or modern interfaces.d with ip:

iface eth0 inet static
    address IP_PRINCIPALE
    netmask 255.255.255.0
    gateway PASSERELLE_PRINCIPALE

iface eth0:0 inet static
    address IP_FAILOVER
    netmask 255.255.255.255

sudo ifup eth0:0

Step 4: Netplan (Ubuntu 22.04+)

/etc/netplan/01-netcfg.yaml (example — adapt):

network:
  version: 2
  ethernets:
    eth0:
      addresses:
        - IP_PRINCIPALE/24
        - IP_FAILOVER/32
      routes:
        - to: default
          via: PASSERELLE_PRINCIPALE
      nameservers:
        addresses: [1.1.1.1, 8.8.8.8]

sudo netplan apply

Step 4b: systemd-networkd

/etc/systemd/network/10-failover.network:

[Match]
Name=eth0

[Network]
Address=IP_PRINCIPALE/24
Address=IP_FAILOVER/32
Gateway=PASSERELLE_PRINCIPALE

sudo systemctl restart systemd-networkd

Services using the failover IP

Nginx vhost

server {
    listen IP_FAILOVER:443 ssl;
    server_name service-failover.tld;
    ...
}

Bind / mail

Configure PTR records in the panel for the failover IP if you send mail.

Firewall

sudo ufw allow from any to IP_FAILOVER port 443 proto tcp

Migration to another server

1. Stop services on the old server.
2. Detach the failover IP in the panel.
3. Attach to the new server.
4. Configure the same IP on the new interface.
5. Verify ARP/propagation (a few minutes max).

Verification

ip addr | grep IP_FAILOVER
curl -4 --interface IP_FAILOVER https://ifconfig.me

From outside:

ping IP_FAILOVER
traceroute IP_FAILOVER

Troubleshooting

| Symptom | Hint |

|----------|-------|

| Ping OK locally, not from Internet | Panel: IP not attached to correct server |

| ARP conflict | Old server still up with same IP |

| Misconfigured /32 | Gateway: usually no gw on failover alias |

Need help?

Ticket with ip addr, panel screenshot (IP attached), server reference, and migration time.