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;
Feedback-ID: i6a694271:Fastmail
Date:   Thu, 13 Oct 2022 11:34:48 -0600
From:   Daniel Xu <dxu@dxuuu.xyz>
To:     Martin KaFai Lau <martin.lau@linux.dev>
Cc:     pablo@netfilter.org, fw@strlen.de, netfilter-devel@vger.kernel.org,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
        andrii@kernel.org, daniel@iogearbox.net, ast@kernel.org,
        bpf@vger.kernel.org, memxor@gmail.com
Subject: Re: [PATCH bpf-next v4 2/3] selftests/bpf: Add connmark read test
Message-ID: <20221013173448.aprptjs5qq777342@k2>
References: <cover.1660254747.git.dxu@dxuuu.xyz>
 <d3bc620a491e4c626c20d80631063922cbe13e2b.1660254747.git.dxu@dxuuu.xyz>
 <43bf4a5f-dac9-4fe9-1eba-9ab9beb650aa@linux.dev>
 <20221012220953.i2xevhu36kxyxscl@k2>
 <16bcda3b-989e-eadf-b6c3-803470b0afd6@linux.dev>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <16bcda3b-989e-eadf-b6c3-803470b0afd6@linux.dev>
Precedence: bulk

On Wed, Oct 12, 2022 at 03:20:01PM -0700, Martin KaFai Lau wrote:
> On 10/12/22 3:09 PM, Daniel Xu wrote:
> > Hi Martin,
> > 
> > On Tue, Oct 11, 2022 at 10:49:32PM -0700, Martin KaFai Lau wrote:
> > > On 8/11/22 2:55 PM, Daniel Xu wrote:
> > > > Test that the prog can read from the connection mark. This test is nice
> > > > because it ensures progs can interact with netfilter subsystem
> > > > correctly.
> > > > 
> > > > Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
> > > > Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
> > > > ---
> > > >    tools/testing/selftests/bpf/prog_tests/bpf_nf.c | 3 ++-
> > > >    tools/testing/selftests/bpf/progs/test_bpf_nf.c | 3 +++
> > > >    2 files changed, 5 insertions(+), 1 deletion(-)
> > > > 
> > > > diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_nf.c b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
> > > > index 88a2c0bdefec..544bf90ac2a7 100644
> > > > --- a/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
> > > > +++ b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
> > > > @@ -44,7 +44,7 @@ static int connect_to_server(int srv_fd)
> > > >    static void test_bpf_nf_ct(int mode)
> > > >    {
> > > > -	const char *iptables = "iptables -t raw %s PREROUTING -j CT";
> > > > +	const char *iptables = "iptables -t raw %s PREROUTING -j CONNMARK --set-mark 42/0";
> > > Hi Daniel Xu, this test starts failing recently in CI [0]:
> > > 
> > > Warning: Extension CONNMARK revision 0 not supported, missing kernel module?
> > >    iptables v1.8.8 (nf_tables): Could not fetch rule set generation id:
> > > Invalid argument
> > > 
> > >    Warning: Extension CONNMARK revision 0 not supported, missing kernel module?
> > >    iptables v1.8.8 (nf_tables): Could not fetch rule set generation id:
> > > Invalid argument
> > > 
> > >    Warning: Extension CONNMARK revision 0 not supported, missing kernel module?
> > >    iptables v1.8.8 (nf_tables): Could not fetch rule set generation id:
> > > Invalid argument
> > > 
> > >    Warning: Extension CONNMARK revision 0 not supported, missing kernel module?
> > >    iptables v1.8.8 (nf_tables): Could not fetch rule set generation id:
> > > Invalid argument
> > > 
> > >    test_bpf_nf_ct:PASS:test_bpf_nf__open_and_load 0 nsec
> > >    test_bpf_nf_ct:FAIL:iptables unexpected error: 1024 (errno 0)
> > > 
> > > Could you help to take a look? Thanks.
> > > 
> > > [0]: https://github.com/kernel-patches/bpf/actions/runs/3231598391/jobs/5291529292
> > 
> > [...]
> > 
> > Thanks for letting me know. I took a quick look and it seems that
> > synproxy selftest is also failing:
> > 
> >      2022-10-12T03:14:20.2007627Z test_synproxy:FAIL:iptables -t raw -I PREROUTING      -i tmp1 -p tcp -m tcp --syn --dport 8080 -j CT --notrack unexpected error: 1024 (errno 2)
> > 
> > Googling the "Could not fetch rule set generation id" yields a lot of
> > hits. Most of the links are from downstream projects recommending user
> > downgrade iptables (nftables) to iptables-legacy.
> 
> Thanks for looking into it!  We have been debugging a bit today also.  I
> also think iptables-legacy is the one to use.  I posted a patch [0].  Let
> see how the CI goes.
> 
> The rules that the selftest used is not a lot.  I wonder what it takes to
> remove the iptables command usage from the selftest?

At least the conntrack mark stuff, it would've been easier to write the
selftests _without_ iptables. But I thought it was both good and
necessary to test interop between BPF and netfilter. B/c that is
what the user is doing (at least for me).

However if it's causing maintenance trouble, I'll leave that call to
you.

Thanks,
Daniel