Initial Network Configuration with OVS

mw123

New Member
Jul 7, 2017
1
0
1
84
Hi, trying to switch over to Proxmox from just using VMM previously. Running into issues on my new install, networking service isn't starting up completely.

When booting up I see a couple of messages saying something to the effect of:
"start job is running for lsb open vswitch switch" and
"start job is running for raise network interfaces"
Meanwhile there is a 5 minute countdown to timeout.

I had a configuration setup similar to example 2 on the Proxmox Open vSwitch page: https://pve.proxmox.com/wiki/Open_vSwitch#Example_2:_Bond_.2B_Bridge_.2B_Internal_Ports.

I found out that I was able to get connectivity by restarting the networking service twice and then enabling the physical interfaces in the bond:

service networking restart
service networking restart
ovs-appctl bond/enable-slave bond0 enp6s0f0
ovs-appctl bond/enable-slave bond0 enp6s0f1

I currently have a copy and pasted version of the above mentioned example 2 as my network config with interfaces and my network specifics modified and still am having the same issues. The fact that restarting the service a couple of times gets the OVS interfaces initialized makes me wonder if there is a start up dependency issue. I am unsure of how to troubleshoot this further.

/etc/network/interfaces:
root@host-0:~# cat /etc/network/interfaces
# Loopback interface
auto lo
iface lo inet loopback

# Bond enp6s0f0 and enp6s0f1 together
allow-vmbr0 bond0
iface bond0 inet manual
ovs_bridge vmbr0
ovs_type OVSBond
ovs_bonds enp6s0f0 enp6s0f1
# Force the MTU of the physical interfaces to be jumbo-frame capable.
# This doesn't mean that any OVSIntPorts must be jumbo-capable.
# We cannot, however set up definitions for enp6s0f0 and enp6s0f1 directly due
# to what appear to be bugs in the initialization process.
pre-up ( ifconfig enp6s0f0 mtu 9000 && ifconfig enp6s0f1 mtu 9000 )
ovs_options bond_mode=balance-tcp lacp=active other_config:lacp-time=fast
mtu 9000

# Bridge for our bond and vlan virtual interfaces (our VMs will
# also attach to this bridge)
auto vmbr0
allow-ovs vmbr0
iface vmbr0 inet manual
ovs_type OVSBridge
# NOTE: we MUST mention bond0, vlan101 even though each
# of them lists ovs_bridge vmbr0! Not sure why it needs this
# kind of cross-referencing but it won't work without it!
ovs_ports bond0 vlan101
mtu 9000

# Proxmox cluster communication vlan
allow-vmbr0 vlan101
iface vlan101 inet static
ovs_type OVSIntPort
ovs_bridge vmbr0
ovs_options tag=101
ovs_extra set interface ${IFACE} external-ids:iface-id=$(hostname -s)-${IFACE}-vif
address 10.16.1.105
netmask 255.255.255.0
gateway 10.16.1.1
mtu 1500
 
In principle the above configuration should work correctly, please note that the line

Code:
pre-up ( ifconfig enp6s0f0 mtu 9000 && ifconfig enp6s0f1 mtu 9000 )

contains "ifconfig" which is not available as default in stretch (from the interface names I conclude you have PVE 5 respectively stretch).
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!