* Hello .. Packet path :)
@ 2003-12-03 15:30 Michael Gale
2003-12-03 16:46 ` Ramin Dousti
0 siblings, 1 reply; 3+ messages in thread
From: Michael Gale @ 2003-12-03 15:30 UTC (permalink / raw)
To: netfilter
Hello,
I am trying to make a packet go through the least amount of chains / tables as possible for performance.
I have read through the online documentation about netfilter and this "Linux Firewalls Second Edition" book (which was ok).
But I still have some questions about the order in which the tables are checked. Here is what I think happens when a pack comes in and should be forwarded to a internal machine
Firewall External interface:
Packet comes in:
NAT table PREROUTING
NAT talbe OUTPUT
NAT table POSTROUTING
filter table INPUT
filter table OUTPUT or forward
Then you would have the same thing when the packet leaves the internal interface.
Of course this is if you break it down by interface first.
Please let me know if this is correct ?
--
Michael Gale
Network Administrator
Utilitran Corporation
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Hello .. Packet path :)
2003-12-03 16:46 ` Ramin Dousti
@ 2003-12-03 15:49 ` Michael Gale
0 siblings, 0 replies; 3+ messages in thread
From: Michael Gale @ 2003-12-03 15:49 UTC (permalink / raw)
To: netfilter
Thank you to all who replied :)
Michael.
On Wed, 3 Dec 2003 11:46:47 -0500
Ramin Dousti <ramin@cannon.eng.us.uu.net> wrote:
> Daniel Chemko had an email on this subject not too long ago:
>
> Raw/PREROUTING
> Mangle/PREROUTING
> Nat/PREROUTING
> Mangle/INPUT
> Filter/INPUT
> Mangle/FORWARD
> Filter/FORWARD
> Raw/OUTPUT
> Mangle/OUTPUT
> Filter/OUTPUT
> Nat/OUTPUT
> Mangle/POSTROUTING
> Nat/POSTROUTING
>
>
> On Wed, Dec 03, 2003 at 08:30:34AM -0700, Michael Gale wrote:
>
> > Hello,
> >
> > I am trying to make a packet go through the least amount of chains / tables as possible for performance.
> >
> > I have read through the online documentation about netfilter and this "Linux Firewalls Second Edition" book (which was ok).
> >
> > But I still have some questions about the order in which the tables are checked. Here is what I think happens when a pack comes in and should be forwarded to a internal machine
> >
> > Firewall External interface:
> > Packet comes in:
> > NAT table PREROUTING
> > NAT talbe OUTPUT
> > NAT table POSTROUTING
> > filter table INPUT
> > filter table OUTPUT or forward
> >
> > Then you would have the same thing when the packet leaves the internal interface.
> >
> > Of course this is if you break it down by interface first.
> >
> > Please let me know if this is correct ?
> >
> > --
> > Michael Gale
> > Network Administrator
> > Utilitran Corporation
>
--
Michael Gale
Network Administrator
Utilitran Corporation
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Hello .. Packet path :)
2003-12-03 15:30 Hello .. Packet path :) Michael Gale
@ 2003-12-03 16:46 ` Ramin Dousti
2003-12-03 15:49 ` Michael Gale
0 siblings, 1 reply; 3+ messages in thread
From: Ramin Dousti @ 2003-12-03 16:46 UTC (permalink / raw)
To: Michael Gale; +Cc: netfilter
Daniel Chemko had an email on this subject not too long ago:
Raw/PREROUTING
Mangle/PREROUTING
Nat/PREROUTING
Mangle/INPUT
Filter/INPUT
Mangle/FORWARD
Filter/FORWARD
Raw/OUTPUT
Mangle/OUTPUT
Filter/OUTPUT
Nat/OUTPUT
Mangle/POSTROUTING
Nat/POSTROUTING
On Wed, Dec 03, 2003 at 08:30:34AM -0700, Michael Gale wrote:
> Hello,
>
> I am trying to make a packet go through the least amount of chains / tables as possible for performance.
>
> I have read through the online documentation about netfilter and this "Linux Firewalls Second Edition" book (which was ok).
>
> But I still have some questions about the order in which the tables are checked. Here is what I think happens when a pack comes in and should be forwarded to a internal machine
>
> Firewall External interface:
> Packet comes in:
> NAT table PREROUTING
> NAT talbe OUTPUT
> NAT table POSTROUTING
> filter table INPUT
> filter table OUTPUT or forward
>
> Then you would have the same thing when the packet leaves the internal interface.
>
> Of course this is if you break it down by interface first.
>
> Please let me know if this is correct ?
>
> --
> Michael Gale
> Network Administrator
> Utilitran Corporation
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2003-12-03 16:46 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-12-03 15:30 Hello .. Packet path :) Michael Gale
2003-12-03 16:46 ` Ramin Dousti
2003-12-03 15:49 ` Michael Gale
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox