Proxmox Host Network Issue (Cannot Ping Gateway)

[clem276]

Description:​

Hi everyone,

I’m experiencing a network issue with my Proxmox host, and I can’t figure out the cause. Below are the details of my setup and the steps I’ve already taken to troubleshoot:

---

Context:​

• Proxmox VE Version: [Specify exact version]
• Host Network Configuration:
  • IP Address: 10.0.1.14
  • Gateway: 10.0.0.1
  • Subnet Mask: /8 (255.0.0.0)
  • Bridge Interface: vmbr0, connected to eno1
• Internet Box Used: Livebox (acting as a gateway with IP 10.0.0.1).
• DHCP Configuration on the Box:
  • DHCP Range: 10.0.1.1 - 10.2.255.255
  • DNS: 10.0.0.1

---

Problem:​

• The Proxmox host cannot ping the gateway (10.0.0.1).
• As a result, the host cannot access the Internet (e.g., pinging 8.8.8.8 or google.com fails).

However:

• VMs connected to the vmbr0 bridge can communicate with the gateway (10.0.0.1) and access the Internet without any issues.
• The gateway is visible in Proxmox’s ARP table but remains in the STALE state.

---

Steps Taken:​

1. Local connectivity check:
   • Proxmox can ping other devices on the local network (10.0.1.x), such as VMs.
2. Traffic capture with tcpdump:
   • ICMP and ARP requests leave Proxmox, but no replies are received from the gateway.
3. ARP entry:
   • I manually added a static ARP entry for 10.0.0.1, but the issue persists.
4. Network configuration verified:
   • The default route is correctly set (default via 10.0.0.1 dev vmbr0).
   • DNS is configured in /etc/resolv.conf with 10.0.0.1.

---

Current Network Configuration on Proxmox:​

Here’s the content of my /etc/network/interfaces file:

auto lo
iface lo inet loopback

auto eno1
iface eno1 inet manual

auto vmbr0
iface vmbr0 inet static
address 10.0.1.14
netmask 255.0.0.0
gateway 10.0.0.1
bridge_ports eno1
bridge_stp off
bridge_fd 0

---

Additional Notes:​

Outputs from the Proxmox Host:​

1. ip addr Output:
   
   • The vmbr0 interface is configured with:
     • IP Address: 10.0.1.14
     • Subnet Mask: /8 (255.0.0.0)
     • MAC Address: b8:2a:72:d8:6a:77
   • The bridge interface vmbr0 is properly linked to eno1.
   • The interface vmbr0 is in the UP state.
2. ip route Output:
   
   • Default route: default via 10.0.0.1 dev vmbr0 proto kernel onlink
   • Subnet route: 10.0.0.0/8 dev vmbr0 proto kernel scope link src 10.0.1.14
3. ip neigh Output:
   
   • The gateway (10.0.0.1) is listed in the ARP table with the MAC address ac:d7:5b:3a:fd:20 but is in the STALE state.
   • Other devices on the network (e.g., 10.0.1.3) are reachable.

Network Behavior Observations:​

• ICMP requests sent to 10.0.0.1 do not receive a response.
• Other devices in the 10.0.1.x range are reachable, including VMs connected via vmbr0.
• VMs can communicate with the gateway and access the Internet without issues.

Configuration Notes:​

1. Network Mask:
   
   • The network mask on Proxmox (/8) matches the Livebox configuration.
   • The Proxmox host uses 10.0.1.14 as its IP and 10.0.0.1 as the gateway.
2. Firewall:
   
   • The Proxmox firewall is currently disabled for testing:
     bash
     CopierModifier
     pve-firewall stop<br>
3. ARP Behavior:
   
   • Proxmox does not seem to receive ARP replies or ICMP echo replies from the gateway (10.0.0.1), as seen in the STALE ARP entry.
4. Traffic Analysis:
   
   • tcpdump on vmbr0 shows ICMP requests leaving the interface, but no replies from the gateway.

---

Questions:​

1. Why is my Proxmox host unable to communicate with the gateway while the VMs can do so without issues?
2. Could this be related to the Livebox configuration blocking traffic from the Proxmox host (10.0.1.14)?
3. Are there any additional steps I should take to debug or resolve this issue?

---

Thank you in advance for your help!

 

[bofh]

probably ip is taken, you assign that ip within the dhcp range. or gateway requires ip by dhcp. certainly a gateway issue