A veth pair has two interfaces one at each end . in this case one of it is df408c5c_eth1, where is the other member of this pair ? thats my question. If we manually create a lxc container , not by, OSA, we need to create a veth pair. the command 'ip link add veth1 type veth peer name veth2' creates two interfaces of the pair as follows:

----

3: veth1@veth2: <BROADCAST,MULTICAST,M-DOWN> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 11:22:33:44:55:77 brd ff:ff:ff:ff:ff:ff
4: veth2@veth1: <BROADCAST,MULTICAST,M-DOWN> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 11:22:33:44:55:88 brd ff:ff:ff:ff:ff:ff

------

as shown above, the two belong to the same pair. but in OSA case, we cant see the other member of df408c5c_eth1. Thats my query. hope you got it....

On Thu, Jan 9, 2025 at 4:58 PM Dmitriy Rabotyagov <noonedeadpunk@gmail.com> wrote:

Hey,

So as other pair of veth is inside of the LXC, you don't really see it
inside of the host system.
I think the easiest way to find which container a visible on host veth
belongs to, is by interface name itself, as the ID you see in
interface name is part of the container name. For instance:
root@control01:~# ip link show type veth | grep df408c5c
41: df408c5c_eth0@if2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500
qdisc noqueue master ovs-system state UP mode DEFAULT group default
qlen 1000
42: df408c5c_eth1@if3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500
qdisc noqueue master ovs-system state UP mode DEFAULT group default
qlen 1000
root@control01:~# lxc-ls -1 | grep df408c5c
control01-placement-container-df408c5c
root@control01:~#

Then you can also check for LXC configuration as mapping of veth is
defined there, ie:
root@control01:~# grep -r df408c5c_eth1 /var/lib/lxc/*/eth1.ini
/var/lib/lxc/control01-placement-container-df408c5c/eth1.ini:lxc.net.1.veth.pair
= df408c5c_eth1
root@control01:~#

Probable there're better ways of doing that, but I'm usually ending up here :)

ср, 8 янв. 2025 г. в 18:24, engineer2024 <engineerlinux2024@gmail.com>:
>
> I have observed that in the OSA setups, on the controller nodes, each lxc container has two networks inside , eth0, eth1 interfaces. Ideally eth1 should be attached to br-mgmt bridge via veth pair. When I do 'ip link show type veth' on the host node, it is listing both the veth interfaces for each lxc container, but it is not listing the other member of that veth pair.
>
> ---
> # ip link show type veth
> 9f9add59_eth0@if2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master lxcbr0 state UP mode DEFAULT group default qlen 1000
> link/ether fe:b6:db:27:e4:a7 brd ff:ff:ff:ff:ff:ff link-netnsid 0
>
> 9f9add59_eth1@if3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br-mgmt state UP mode DEFAULT group default qlen 1000
> link/ether fe:a8:30:7a:ac:3f brd ff:ff:ff:ff:ff:ff link-netnsid 0
> ---
> ---
> # ip link show type veth | grep 9f9add59
> 72: 9f9add59_eth0@if2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master lxcbr0 state UP mode DEFAULT group default qlen 1000
> 73: 9f9add59_eth1@if3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br-mgmt state UP mode DEFAULT group default qlen 1000
> #
> ---
>
> But I can't find the other member of the veth pair, for example, for the interface 9f9add59_eth1, in the ip list. But the containers are running and reachable on the br-mgmt network. How to find out the other end of a veth pair, which is connected to the bridge ?