<div dir="ltr">Hi,<div>You can set MTU on the network level, see:</div><div><a href="https://docs.openstack.org/api-ref/network/v2/index.html?expanded=create-network-detail#create-network">https://docs.openstack.org/api-ref/network/v2/index.html?expanded=create-network-detail#create-network</a><br></div><div><br></div><div>Lajos Katona (lajoskatona)</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">shubjero <<a href="mailto:shubjero@gmail.com">shubjero@gmail.com</a>> ezt írta (időpont: 2021. szept. 13., H, 16:28):<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hey Openstackers,<br>
<br>
I am trying to determine how to configure/manipulate an OpenStack<br>
instances MTU configuration upon deployment/creation. I know that in<br>
neutron.conf you can set the 'global_physnet_mtu = value' and<br>
'path_mtu = value' in ml2_conf.ini. In our case we are setting these<br>
to 9000 as the physical interfaces of our control plane and compute<br>
nodes are all using 9000 mtu however the bonded interfaces for our<br>
various networks are 8900.<br>
<br>
With this mtu configured, we've noticed that our OpenStack neutron<br>
networks within a tenancy are auto-selecting a mtu of 8958 as from<br>
what I've read it will take the 9000 mtu and minus the approriate<br>
overhead to pass the underlying physical network properly but I think<br>
because our bonded interfaces are 8900 this has started to introduce<br>
some problems in instances with operating systems that hard code the<br>
MTU from Neutron networks.<br>
<br>
When an instance gets deployed it may (Ubuntu 20.04) or may not<br>
(Ubuntu 18.04) retrieve the network mtu from neutron metadata service<br>
and builds it in to the netplan configuration as provided below.<br>
<br>
My question is, where in Neutron could I configure a lower MTU for the<br>
instances? I am not sure if I should change my global_physnet_mtu and<br>
path_mtu to 8900 OR some other configuration option? Thanks all!<br>
<br>
----<br>
<br>
# curl --silent<br>
<a href="http://169.254.169.254/openstack/2018-08-27/network_data.json" rel="noreferrer" target="_blank">http://169.254.169.254/openstack/2018-08-27/network_data.json</a> | jq .<br>
{<br>
"links": [<br>
{<br>
"id": "tap4318b50a-76",<br>
"vif_id": "4318b50a-7685-4777-917f-eea947aa1cb6",<br>
"type": "ovs",<br>
"mtu": 8958, <--- Matches the MTU of the underlying OpenStack<br>
network attached to this instance<br>
"ethernet_mac_address": "fa:16:3e:a4:19:44"<br>
}<br>
],<br>
"networks": [<br>
{<br>
"id": "network0",<br>
"type": "ipv4_dhcp",<br>
"link": "tap4318b50a-76",<br>
"network_id": "4acb077c-0656-4731-ba42-031a22af7931"<br>
}<br>
],<br>
"services": []<br>
}<br>
<br>
----<br>
<br>
# cat /etc/netplan/50-cloud-init.yaml<br>
# This file is generated from information provided by the datasource. Changes<br>
# to it will not persist across an instance reboot. To disable cloud-init's<br>
# network configuration capabilities, write a file<br>
# /etc/cloud/cloud.cfg.d/99-disable-network-config.cfg with the following:<br>
# network: {config: disabled}<br>
network:<br>
version: 2<br>
ethernets:<br>
ens3:<br>
dhcp4: true<br>
match:<br>
macaddress: fa:16:3e:a4:19:44<br>
mtu: 8958 <--- MTU gets hard-coded in the config for<br>
Ubuntu 20.04 but not Ubuntu 18.04.<br>
set-name: ens3<br>
<br>
</blockquote></div>