All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vladimir Nikishkin <lockywolf@gmail.com>
To: Martin Gignac <martin.gignac@gmail.com>
Cc: netfilter@vger.kernel.org
Subject: Re: Matching metainformation cgroup fails on input, works on output.
Date: Wed, 08 Dec 2021 22:17:41 +0800	[thread overview]
Message-ID: <8735n32nax.fsf@laptop.lockywolf.net> (raw)
In-Reply-To: <CANf9dFMcpnqO-at25EKfxad7fX=yT9s0A7EDBgac1doL3fGxpg@mail.gmail.com>

I have just tested it (again) by flushing the old ruleset and using the
following ruleset:

```
table ip filter {
	counter test-icmp-output {
		packets 3 bytes 252
	}

	counter test-icmp-input {
		packets 0 bytes 0
	}

	chain INPUT {
		type filter hook input priority filter; policy accept;
		meta cgroup != 4096 ip saddr 8.8.8.8 ip protocol icmp counter name "test-icmp-input"
	}

	chain OUTPUT {
		type filter hook output priority filter; policy accept;
		meta cgroup != 4096 ip daddr 8.8.8.8 ip protocol icmp counter name "test-icmp-output"
	}
}
```

As previously, the cgroup 0x001000 does not exist. The three outbound
packets are the three ping packets, and they were successfully replied
to.

Martin Gignac <martin.gignac@gmail.com> writes:

> What is the complete output of 'nft list ruleset'?
>
> Is it possible you have an earlier INPUT rule that matches and allows
> packets that match connection-tracking "established" state, such as:
>
>         chain INPUT {
>                 type filter hook input priority filter; policy drop;
>                 ct state established,related counter packets 391638047
> bytes 93651768866 accept
>                 ct state invalid drop
>                 [...]
>
> -Martin
>
>
> On Wed, Dec 8, 2021 at 4:39 AM Vladimir Nikishkin <lockywolf@gmail.com> wrote:
>>
>> Hello, everyone.
>>
>> I have a weird problem!
>>
>> This is my nft code:
>>
>> ```
>> nft add counter filter test-icmp-output
>> nft add counter filter test-icmp-input
>> nft add rule filter OUTPUT meta cgroup != 0x001000 ip daddr 8.8.8.8 ip protocol icmp counter name test-icmp-output
>> nft add rule filter INPUT  meta cgroup != 0x001000 ip saddr 8.8.8.8 ip protocol icmp counter name test-icmp-input
>> ```
>>
>> Pinging 8.8.8.8 works. The packets are visible on tcpdump too.
>> The cgroup id 0x001000 does not exist, so every packet should match.
>>
>> Still, the output counter counts the expected number of packets, the
>> second stays 0.
>>
>> What am I doing wrong?
>>
>> --
>> Your sincerely,
>> Vladimir Nikishkin (MiEr, lockywolf)
>> (Laptop)


-- 
Your sincerely,
Vladimir Nikishkin (MiEr, lockywolf)
(Laptop)

  reply	other threads:[~2021-12-08 14:17 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-08  9:07 Matching metainformation cgroup fails on input, works on output Vladimir Nikishkin
2021-12-08 14:07 ` Martin Gignac
2021-12-08 14:17   ` Vladimir Nikishkin [this message]
2021-12-08 14:51     ` Martin Gignac

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8735n32nax.fsf@laptop.lockywolf.net \
    --to=lockywolf@gmail.com \
    --cc=martin.gignac@gmail.com \
    --cc=netfilter@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.