@olivierlambert Sure!

By way of example, when a new XCP-ng host is setup, all interfaces are created with an MTU of 1500 (even 10G interfaces).

If I create a bond interface on top of two of the 10G interfaces, the default MTU is set to 1500.

If I then create a VLAN interface on top of the bond interface, and set the MTU to 9000, XO happily lets me do so, but jumbo packets over it don't work because the MTU on the underlying interfaces is still 1500 (e.g. ping -s 1700 via the new interface fails).

If I destroy and re-create the underlying interface (bond0) with an MTU of 9000, then re-create the VLAN interface on top with an MTU of 9000, then things work much better (ping -s 1700 via the new interface succeeds).

Does this help?

Thanks

@olivierlambert Sure!

By way of example, when a new XCP-ng host is setup, all interfaces are created with an MTU of 1500 (even 10G interfaces).

If I create a bond interface on top of two of the 10G interfaces, the default MTU is set to 1500.

If I then create a VLAN interface on top of the bond interface, and set the MTU to 9000, XO happily lets me do so, but jumbo packets over it don't work because the MTU on the underlying interfaces is still 1500 (e.g. ping -s 1700 via the new interface fails).

If I destroy and re-create the underlying interface (bond0) with an MTU of 9000, then re-create the VLAN interface on top with an MTU of 9000, then things work much better (ping -s 1700 via the new interface succeeds).

Does this help?

Thanks

Hi,

I've appreciated the things that can be done through the GUI in the XOA interface, but when creating new interfaces (bonded, vlan, etc.) in XO, it would be nice if the appliance validated that the underlying MTU is at least as large as the MTU of the new interface being created on top of it.

This is a low-priority feature request.

Thanks,

• Daniel