Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230711043453.64095-1-ivan@cloudflare.com> <20230711193612.22c9bc04@kernel.org>
In-Reply-To: <20230711193612.22c9bc04@kernel.org>
From:   Yan Zhai <yan@cloudflare.com>
Date:   Wed, 12 Jul 2023 11:42:26 -0500
Message-ID: <CAO3-PbrZHn1syvhb3V57oeXigE_roiHCbzYz5Mi4wiymogTg2A@mail.gmail.com>
Subject: Re: [RFC PATCH net-next] tcp: add a tracepoint for tcp_listen_queue_drop
To:     Jakub Kicinski <kuba@kernel.org>
Cc:     Ivan Babrou <ivan@cloudflare.com>, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org, kernel-team@cloudflare.com,
        Eric Dumazet <edumazet@google.com>,
        "David S. Miller" <davem@davemloft.net>,
        Paolo Abeni <pabeni@redhat.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Masami Hiramatsu <mhiramat@kernel.org>,
        David Ahern <dsahern@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Jul 11, 2023 at 9:36=E2=80=AFPM Jakub Kicinski <kuba@kernel.org> wr=
ote:
>
> On Mon, 10 Jul 2023 21:34:52 -0700 Ivan Babrou wrote:
> > There's already a way to count the overall numbers of queue overflows:
> >
> >     $ sudo netstat -s | grep 'listen queue'
> >     4 times the listen queue of a socket overflowed
> >
> > However, it's too coarse for monitoring and alerting when a user wants =
to
> > track errors per socket and route alerts to people responsible for thos=
e
> > sockets directly. For UDP there's udp_fail_queue_rcv_skb, which fills
> > a similar need for UDP sockets. This patch adds a TCP equivalent.
>
> Makes me want to revert your recent UDP tracepoint to be honest :(
> We can play whack a mole like this. You said that kfree_skb fires
> too often, why is that? Maybe it's an issue of someone using
> kfree_skb() when they should be using consume_skb() ?

Hi Jakub,

  The issue with kfree_skb is not that it fires too frequently (not in
the 6.x kernel now). Rather, it is unable to locate the socket info
when a SYN is dropped due to the accept queue being full. The sk is
stolen upon inet lookup, e.g. in tcp_v4_rcv. This makes it unable to
tell in kfree_skb which socket a SYN skb is targeting (when TPROXY or
socket lookup are used). A tracepoint with sk information will be more
useful to monitor accurately which service/socket is involved.

--
Yan