Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
User-Agent: Cyrus-JMAP/3.5.0-alpha0-403-gbc3c488b23-fm-20210419.005-gbc3c488b
Mime-Version: 1.0
Message-Id: <52f1b7a1-ff3b-41d3-a84b-badcda8a6ad6@www.fastmail.com>
In-Reply-To: <8e0aa5a6-0457-ccd0-8984-9c9aaeab2228@gmail.com>
References: <d3a4afd7-a448-4310-930d-063b39bde86e@www.fastmail.com>
 <8e0aa5a6-0457-ccd0-8984-9c9aaeab2228@gmail.com>
Date:   Mon, 26 Apr 2021 23:34:55 -0300
From:   "Henrique de Moraes Holschuh" <hmh@hmh.eng.br>
To:     "Eric Dumazet" <eric.dumazet@gmail.com>,
        linux-kernel@vger.kernel.org, netdev@vger.kernel.org
Subject: Re: Unexpected timestamps in tcpdump with veth + tc qdisc netem delay
Content-Type: text/plain
Precedence: bulk

On Mon, Apr 26, 2021, at 14:07, Eric Dumazet wrote:
> On 4/26/21 4:35 PM, Henrique de Moraes Holschuh wrote:
[...]
> > [root netns]: tcpdump -i vec0 -s0 -n -p net 192.168.233.0/30
> > listening on vec0, link-type EN10MB (Ethernet), capture size 262144 bytes
> > 17:09:09.740681 IP 192.168.233.1 > 192.168.233.2: ICMP echo request, id 9327, seq 1, length 64
> 
> Here you see the packet _after_ the 250ms delay
> > 17:09:09.990891 IP 192.168.233.2 > 192.168.233.1: ICMP echo reply, id 9327, seq 1, length 64
> Same here.
[...]
> > Adding more namespaces and VETH pairs + routing "in a row" so that the packet "exits" one veth tunnel and enters another one (after trivial routing) doesn't fix the tcpdump timestamps in the capture at the other end of the veth-veth->routing->veth-veth->routing->... chain.
> > 
> > It looks like some sort of bug to me, but maybe I am missing something, in which case I would greatly appreciate an explanation of where I went wrong... 
> 
> That is only because you expect to see something, but you forgot that tcpdump captures TX packet _after_ netem.

That was it!  Thank you very much for the quick reply, and direct, precise explanation.

I had completely forgotten about the qdisc running before interface timestamping, and I overlooked the fact that I did not try three netns in-a-chain with the qdiscs *in the middle netns*: I tried them in the two opposite edge netns, only.

Again, thank you!

-- 
  Henrique Holschuh