netfilter-devel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Patrick McHardy <kaber@trash.net>
To: "Bjørnar Ness" <bjornar.ness@gmail.com>
Cc: netfilter-devel@vger.kernel.org
Subject: Re: nftables and sets
Date: Wed, 22 Jan 2014 15:26:01 +0000	[thread overview]
Message-ID: <20140122152600.GB14817@macbook.localnet> (raw)
In-Reply-To: <CAJO99T=HBk7_PJG8+SmkqA7NUDuWBvujGckBJ=R_wn0i-7MOsA@mail.gmail.com>

[ please don't drop CCs ]

On Wed, Jan 22, 2014 at 04:09:15PM +0100, Bjørnar Ness wrote:
> 2014/1/22 Patrick McHardy <kaber@trash.net>:
> > On Wed, Jan 22, 2014 at 03:09:30PM +0100, Bjørnar Ness wrote:
> >> Is it planned (or already implemented) support for multiple value vmaps?
> >> for example {type uid, type ipv4 : verdict}
> >
> > We support so called concatenations, basically multiple types glued
> > together. You can use those for exact matches. However the kernel side
> > is missing a few bits so far.
> >
> > Basically:
> >
> > nft add map uidmap { type uid . ipv4_address : verdict; }
> > nft filter input uid . ip saddr @uidmap
> > nft add element uidmap { root . localhost : drop }
> 
> Ok, I see.. this sortoff works, any plans to change from concat to
> multiple types?

Why? It is multiple types. The thing is if you do multidimensional
lookups things get a lot more inefficient. For the purpose of
equality (or membership) it doesn't make any difference.

> > The combination doesn't make too much sense of course.
> >
> >> Also, using iptables I am doing something like this
> >>
> >> -A INPUT -j BLOCK
> >> -A BLOCK -m set --match-set ignorelist src -j RETURN
> >> -A BLOCK -m set --match-set blocklist src -j DROP
> >>
> >> ..to make sure ips listed in ignorelist does not get blocked..
> >>
> >> How would I do the same using nft?
> >
> > nft add set ignorelist { type ipv4_address; }
> > nft add set blocklist { type ipv4_address; }
> >
> > nft filter input jump block
> > nft filter block ip saddr @ignorelist return
> > nft filter block ip saddr @blocklist drop
> 
> Could this be done in cleaner way with nft, for example
> 
> nft filter input ip saddr @ignorelist jump_over_next_rule_or_to_label
> nft filter input ip saddr @blocklist drop
> nft filter input add label jumplabel

I guess we have different opinions what constitutes cleaner.
We don't support jumps to rules, but it should be possible
quite easily to change the code to support:

nft filter input ip saddr !@ignorelist ip saddr @blocklist drop

to put this into a single rule.
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

      parent reply	other threads:[~2014-01-22 15:26 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-22 14:09 nftables and sets Bjørnar Ness
2014-01-22 14:44 ` Patrick McHardy
     [not found]   ` <CAJO99T=HBk7_PJG8+SmkqA7NUDuWBvujGckBJ=R_wn0i-7MOsA@mail.gmail.com>
2014-01-22 15:26     ` Patrick McHardy [this message]

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=20140122152600.GB14817@macbook.localnet \
    --to=kaber@trash.net \
    --cc=bjornar.ness@gmail.com \
    --cc=netfilter-devel@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).