All of lore.kernel.org
 help / color / mirror / Atom feed
* Simple question about ipset
@ 2006-09-20  6:54 Ming-Ching Tiew
  0 siblings, 0 replies; 4+ messages in thread
From: Ming-Ching Tiew @ 2006-09-20  6:54 UTC (permalink / raw)
  To: netfilter


I have been reading ipset from http://ipset.netfilter.org/features.html
quite a few times but still do not understand what does it mean
by this :-

    iptables -A FORWARD -m set --set servers dst,dst -j ACCEPT

My question is why the flag is dst,dst ?

And similarly what is the significance if it is src,dst ?
What if it is src,dst,dst and so on ?

Regards






^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Simple question about ipset
@ 2006-09-20 13:38 Ming-Ching Tiew
  2006-09-20 13:57 ` Jozsef Kadlecsik
  0 siblings, 1 reply; 4+ messages in thread
From: Ming-Ching Tiew @ 2006-09-20 13:38 UTC (permalink / raw)
  To: netfilter



>I have been reading ipset from http://ipset.netfilter.org/features.html
>quite a few times but still do not understand what does it mean
>by this :-
>
>    iptables -A FORWARD -m set --set servers dst,dst -j ACCEPT
>
>My question is why the flag is dst,dst ?
>
>And similarly what is the significance if it is src,dst ?
>What if it is src,dst,dst and so on ?

Gosh I think something sudden sparked my mind, I think I understand it now.
The number of flags ties to the bindings, ie this example it is ip->port, 
then dst,dst would means check the destination ip, and destination port.
Had the flags be src,dst, then it is checking the source ip and destination port.

I hope my understanding is correct. Perhaps the docs should explain it more 
clearly.

Cheers. 




^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Simple question about ipset
  2006-09-20 13:38 Simple question about ipset Ming-Ching Tiew
@ 2006-09-20 13:57 ` Jozsef Kadlecsik
  2006-09-24 20:34   ` Ming-Ching Tiew
  0 siblings, 1 reply; 4+ messages in thread
From: Jozsef Kadlecsik @ 2006-09-20 13:57 UTC (permalink / raw)
  To: Ming-Ching Tiew; +Cc: netfilter

On Wed, 20 Sep 2006, Ming-Ching Tiew wrote:

> >I have been reading ipset from http://ipset.netfilter.org/features.html
> >quite a few times but still do not understand what does it mean
> >by this :-
> >
> >    iptables -A FORWARD -m set --set servers dst,dst -j ACCEPT
> >
> >My question is why the flag is dst,dst ?
> >
> >And similarly what is the significance if it is src,dst ?
> >What if it is src,dst,dst and so on ?
>
> Gosh I think something sudden sparked my mind, I think I understand it now.
> The number of flags ties to the bindings, ie this example it is ip->port,
> then dst,dst would means check the destination ip, and destination port.
> Had the flags be src,dst, then it is checking the source ip and destination port.

Exactly. Also, if the set type itself store data pairs like in the case of
ipporthash, you have to define what kind of info from the packet must be
matched against the given set: (source or dest ip) and (source or dest)
port. I.e if set 'servers' is of type ipporthash, you should use

iptables -A FORWARD -m set --set servers dst,dst -j ACCEPT

to match all of your servers (IP addresses) and their service ports, if
the set is properly filled up.

> I hope my understanding is correct. Perhaps the docs should explain it more
> clearly.

Yes, the docs are terse. Patches against the docs are (also) always
welcomed.

Best regards,
Jozsef
-
E-mail  : kadlec@blackhole.kfki.hu, kadlec@sunserv.kfki.hu
PGP key : http://www.kfki.hu/~kadlec/pgp_public_key.txt
Address : KFKI Research Institute for Particle and Nuclear Physics
          H-1525 Budapest 114, POB. 49, Hungary


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Simple question about ipset
  2006-09-20 13:57 ` Jozsef Kadlecsik
@ 2006-09-24 20:34   ` Ming-Ching Tiew
  0 siblings, 0 replies; 4+ messages in thread
From: Ming-Ching Tiew @ 2006-09-24 20:34 UTC (permalink / raw)
  To: netfilter

From: "Jozsef Kadlecsik" <kadlec@blackhole.kfki.hu>
> 
> > I hope my understanding is correct. Perhaps the docs should explain it more
> > clearly.
> 
> Yes, the docs are terse. Patches against the docs are (also) always
> welcomed.
> 

If my understanding is correct ( which I am still not very sure at this moment ),
perhaps may I suggest the syntax be changed such that only one flag is 
allowed ( and necessary ) and the value can be 'dst', 'src', or 'both' ?

  eg,
             iptables -A FORWARD -m set --set servers dst -j ACCEPT
             iptables -A FORWARD -m set --set servers src -j ACCEPT
             iptables -A FORWARD -m set --set servers both -j ACCEPT

                or if the keyword 'both' is left out, it is then implied :-
 
           iptables -A FORWARD -m set --set servers -j ACCEPT

The nature of relationship is that it will be hard to form a meaningful
relationship where the flags could be heterogeneous, so why don't
just implement a simplied syntax ?

Cheers

 




^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2006-09-24 20:34 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-09-20 13:38 Simple question about ipset Ming-Ching Tiew
2006-09-20 13:57 ` Jozsef Kadlecsik
2006-09-24 20:34   ` Ming-Ching Tiew
  -- strict thread matches above, loose matches on Subject: below --
2006-09-20  6:54 Ming-Ching Tiew

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.