Arch Linux

Maniaxx

Member

Registered: 2014-05-14

Posts: 738

[SOLVED]Network bridge eth0/tap0 with iproute2

Hi,
i want to set up a network bridge with eth0 and tap0 (for QEMU) manually with iproute2 (no deprecated tools (like brctl), no QEMU bridge-helper or scripts).

It results in no layer3 traffic. Wireshark shows ARP requests to the gateway in both directions on eth0 (and br0 only incoming). I also cannot ping the gateway (192.168.1.1). 100% packet loss.

I want internet access for eth0/tap0 and they need to be able to communicate to each other. I assigned different IPs and MACs to the NICs. I'm not sure if that's correct as some examples just flush the IP on eth0. But if i remove the ip from eth0 i'm not sure how to contact the host from tap0 (QEMU).

Anyone know what's wrong here?

This is my script:

#!/bin/bash

if [[ ! $(grep br0 < /proc/net/dev) ]]; then
	echo "Creating bridge"
	systemctl stop NetworkManager.service
	ip tuntap add tap0 mode tap user 1000
	ip link set tap0 up
	ip link add dev br0 type bridge
	ip link set eth0 promisc on
	ip link set eth0 master br0
	ip link set tap0 master br0
	sysctl net.ipv4.ip_forward=1 >/dev/null 2>&1
	ip addr add 192.168.1.7/24 broadcast 192.168.1.255 dev tap0
	ip addr add 192.168.1.8/24 broadcast 192.168.1.255 dev br0
	ip link set br0 address 00:25:22:11:11:11
	ip link set dev br0 up
else
	echo "Removing bridge"
	ip link set eth0 nomaster
	ip link set tap0 nomaster
	ip link set eth0 down
	ip tuntap del tap0 mode tap
	ip addr flush dev br0
	ip link del dev br0 type bridge
	ip link set eth0 promisc off
	ip link set eth0 up
	ip route add default via 192.168.1.1 dev eth0 proto static metric 100
	sysctl net.ipv4.ip_forward=0 >/dev/null 2>&1
	systemctl start NetworkManager.service
fi

$ ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc fq_codel master br0 state UP group default qlen 1000
    link/ether 00:25:22:33:33:33 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.6/24 brd 192.168.1.255 scope global eth0
       valid_lft forever preferred_lft forever
61: tap0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc fq_codel master br0 state DOWN group default qlen 1000
    link/ether ae:41:d6:f3:1c:f6 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.7/24 brd 192.168.1.255 scope global tap0
       valid_lft forever preferred_lft forever
62: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 00:25:22:11:11:11 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.8/24 brd 192.168.1.255 scope global br0
       valid_lft forever preferred_lft forever
    inet6 fe80::225:22ff:fe11:1111/64 scope link 
       valid_lft forever preferred_lft forever

$ ip route
default via 192.168.1.1 dev eth0 proto static metric 100 
192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.6 
192.168.1.0/24 dev tap0 proto kernel scope link src 192.168.1.7 linkdown 
192.168.1.0/24 dev br0 proto kernel scope link src 192.168.1.8

Last edited by Maniaxx (2017-11-09 06:02:10)

---

sys2064

nesk

Member

Registered: 2011-03-31

Posts: 181

Re: [SOLVED]Network bridge eth0/tap0 with iproute2

I want internet access for eth0/tap0 and they need to be able to communicate to each other. I assigned different IPs and MACs to the NICs. I'm not sure if that's correct as some examples just flush the IP on eth0. But if i remove the ip from eth0 i'm not sure how to contact the host from tap0 (QEMU).
Anyone know what's wrong here?

Remove the IP from eth0 - packets from tap0 will reach the bridge's IP.
Remove the IP from tap0 as well. TAP provides link layer interface (Ethernet) - the guest is supposed to use it with IP addressing and so on.
Here's how things look like with my setup (eno1 is physical interface) - with VM booted up and able to access the internet:

2: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bridge0 state UP group default qlen 1000
    link/ether 11:22:33:44:44:55 brd ff:ff:ff:ff:ff:ff
5: bridge0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 33:11:22:55:77:88 brd ff:ff:ff:ff:ff:ff
    inet 192.168.0.49/24 brd 192.168.0.255 scope global bridge0
       valid_lft forever preferred_lft forever
6: tap0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bridge0 state UNKNOWN group default qlen 1000
    link/ether 66:77:88:99:22:11 brd ff:ff:ff:ff:ff:ff

i want to set up a network bridge with eth0 and tap0 (for QEMU) manually with iproute2 (no deprecated tools (like brctl), no QEMU bridge-helper or scripts).

Why? Educational experience? All this can be accomplished with NetworkManager (which you apparently have, judging from your scripts) managing the bridge(s) and qemu-bridge-helper creating taps. Works well for me with 2 bridges and several VMs.

Maniaxx

Member

Registered: 2014-05-14

Posts: 738

Re: [SOLVED]Network bridge eth0/tap0 with iproute2

Why? Educational experience?

Yes, i want to do it manually at least once.

Thanks so far. It does work now.

They say traffic is queued twice and gateway traffic can be force routed to eth0 but i couldn't get it completely done.
http://ebtables.netfilter.org/examples/ … l#ex_speed
http://ebtables.netfilter.org/examples/ … x_redirect
http://ebtables.netfilter.org/examples/ … l#example2 (scroll down to "Acknowledgements, Final Comments and Useful Links")

with this:

ebtables -t broute -A BROUTING -d ${MAC_ETH0} -p ipv4 -j redirect --redirect-target DROP
ebtables -t broute -A BROUTING -p ARP -i eth0 -d ${MAC_ETH0} -j DROP

Keeping eth0 IP intact (no br0 IP, different br0 MAC).
It does work but i cannot ping eth0 from tap0/QEMU. Both have internet access though and eth0 can ping tap0 IP. Maybe needs an exception for tap0 somewhere.

---

sys2064