Why does vxlan_sys_4789 have a large MTU size

Solution In Progress - Updated -

Environment

  • OpenShift Container Platform 3

Issue

  • We have the vxlan configured with an mtu of 1450 in the node configmaps for our servers. On ifconfig, we see a device called vxlan_sys_4789, which has a much larger MTU. Is this expected?

Resolution

  • The large MTU is configured as expected as this interface acts as a tunnel between OCP nodes

Root Cause

  • This interface encapsulates traffic with vxlan protocol so the packet going through the tunnel can be larger than the same packet once it's de-encapsulated, but since the vxlan encapsulates traffic generated in other interfaces (with "normal" mtu's) this isn't a problem

Diagnostic Steps

# ip addr show vxlan_sys_4789
123: vxlan_sys_4789: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 65000 qdisc noqueue master ovs-system state UNKNOWN group default qlen 1000
    link/ether 52:db:42:7b:2b:50 brd ff:ff:ff:ff:ff:ff
    inet6 fe80::50db:42ff:fe7b:2b50/64 scope link 
       valid_lft forever preferred_lft forever

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.

Comments