Re: [PATCH net-next,v5 0/9] netfilter: flowtable bridge and vlan enhancements

[Alexander Lobakin]

From: Pablo Neira Ayuso <pablo@xxxxxxxxxxxxx>
Date: Sun, 22 Nov 2020 12:42:19 +0100

> On Sun, Nov 22, 2020 at 10:26:16AM +0000, Alexander Lobakin wrote:
>> From: Pablo Neira Ayuso <pablo@xxxxxxxxxxxxx>
>> Date: Fri, 20 Nov 2020 13:49:12 +0100
> [...]
>>> Something like this:
>>>
>>>                        fast path
>>>                 .------------------------.
>>>                /                          \
>>>                |           IP forwarding   |
>>>                |          /             \  .
>>>                |       br0               eth0
>>>                .       / \
>>>                -- veth1  veth2
>>>                    .
>>>                    .
>>>                    .
>>>                  eth0
>>>            ab:cd:ef:ab:cd:ef
>>>                   VM
>>
>> I'm concerned about bypassing vlan and bridge's .ndo_start_xmit() in
>> case of this shortcut. We'll have incomplete netdevice Tx stats for
>> these two, as it gets updated inside this callbacks.
>
> TX device stats are being updated accordingly.
>
> # ip netns exec nsr1 ip -s link
> 1: lo: <LOOPBACK> mtu 65536 qdisc noop state DOWN mode DEFAULT group default qlen 1000
>     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
>     RX: bytes  packets  errors  dropped overrun mcast   
>     0          0        0       0       0       0       
>     TX: bytes  packets  errors  dropped carrier collsns 
>     0          0        0       0       0       0       
> 2: veth0@if2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
>     link/ether 82:0d:f3:b5:59:5d brd ff:ff:ff:ff:ff:ff link-netns ns1
>     RX: bytes  packets  errors  dropped overrun mcast   
>     213290848248 4869765  0       0       0       0       
>     TX: bytes  packets  errors  dropped carrier collsns 
>     315346667  4777953  0       0       0       0       
> 3: veth1@if2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
>     link/ether 4a:81:2d:9a:02:88 brd ff:ff:ff:ff:ff:ff link-netns ns2
>     RX: bytes  packets  errors  dropped overrun mcast   
>     315337919  4777833  0       0       0       0       
>     TX: bytes  packets  errors  dropped carrier collsns 
>     213290844826 4869708  0       0       0       0       
> 4: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP mode DEFAULT group default qlen 1000
>     link/ether 82:0d:f3:b5:59:5d brd ff:ff:ff:ff:ff:ff
>     RX: bytes  packets  errors  dropped overrun mcast   
>     4101       73       0       0       0       0       
>     TX: bytes  packets  errors  dropped carrier collsns 
>     5256       74       0       0       0       0       

Aren't these counters very low for br0, despite that br0 is an
intermediate point of traffic flow?

> 5: veth0.10@veth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br0 state UP mode DEFAULT group default qlen 1000
>     link/ether 82:0d:f3:b5:59:5d brd ff:ff:ff:ff:ff:ff
>     RX: bytes  packets  errors  dropped overrun mcast   
>     4101       73       0       0       0       62      
>     TX: bytes  packets  errors  dropped carrier collsns 
>     315342363  4777893  0       0       0       0