Hosts or VM's not reaching or not being reachable to/from some IP's

search cancel

book

calendar_today

VMware vSphere ESXi

ARP requests being sent to a vmkernel (vmk) or VM, but no ARP replies are received.
When primary firewall fails over so that secondary firewall is active (with the same virtual IP as before), ESXi hosts are losing Management Network connectivity or not reachable from vCenter.
The ESXi host vmk interface does not resolve the MAC address for the gateway IP on their arp table:
```
esxcli network ip neighbor list
```
ARP requests are sent upstream from the host on one team uplink, but the ARP reply returns on a different uplink.
ARP requests leave the virtual machine, but no ARP reply returns.
The VM ARP Cache has no entry for the destination MAC Address.
```
arp -a
```

VMware vSphere ESXi

This issue occurs due to a misconfiguration in the virtual network or the underlying physical network infrastructure. Common causes include:

Load Balancing Mismatch: The port group uses Route Based on Originating Virtual Port, while the connected physical switch ports are configured with a static Link Aggregation Group / Port Channel (mode on).
VLAN Trunking Issue: Incorrect or missing VLAN trunking configuration on the upstream physical switches.

It is advised to review the networking configuration and engage the physical network infrastructure team for further analysis such as:

Review the Nic Teaming configuration between the ESXi host and the Physical switches.
Review the VLAN Trunking configuration between the ESXi host and the Physical switches.

Note: In certain scenarios, if Etherchannels are being used, you might need to set the impacted portgroup load balancing policy to Route Based on IP Hash or change to Inherit from vSwitch if the virtual switch load balancing is already set to Route Based on IP Hash. For more information, review Configure NIC Teaming, Failover, and Load Balancing on a vSphere Standard Switch or Standard Port Group.

thumb_up Yes

thumb_down No