[PATCH] patch-o-matic-ng reorganization and updated patches

[PATCH] patch-o-matic-ng reorganization and updated patches

[Jozsef Kadlecsik]

Hello,

A lot of modification happened today in pom-ng:

- last two Netfilter_POM.pm patches are applied
- configure.help files are renamed as
  linux/Documentation/Configure.help.ladd and Netfilter_POM.pm
  is changed to take the change into account
- the 2.6 port of the nf-log, raw and TRACE patches are added
  to pom-ng

Upcoming modifications:

- relax the version checking, because it's too rigid: foo-1.2.patch
  should support anything equal and above 1.2 if there is no better
  match
- 2.6 ports of tcp-window-tracking and conntrack_foo patches

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

tcp-window-tracking for 2.6? (Re: [PATCH] patch-o-matic-ng reorganization and updated patches)

[Nuutti Kotivuori]

Jozsef Kadlecsik wrote:
> Upcoming modifications:

[...]

> - 2.6 ports of tcp-window-tracking and conntrack_foo patches

I am personally very interested in the state of the
tcp-window-tracking patch, especially for 2.6 kernels. It is very nice
to know it is being ported for 2.6.

What I would be doing would either be a patch that depends on the
tcp-window-tracking patch to work at all - or it would do something
similar itself which would most probably make it conflict with the
tcp-window-tracking patch.

Does anyone know of the status it? How far is it from being integrated
into the mainline netfilter? Are there any blocking issues why it
cannot be or will not be integrated?

Thanks,
-- Naked

Re: tcp-window-tracking for 2.6? (Re: [PATCH] patch-o-matic-ng reorganization and updated patches)

[Jozsef Kadlecsik]

On Thu, 19 Feb 2004, Nuutti Kotivuori wrote:

> > - 2.6 ports of tcp-window-tracking and conntrack_foo patches
>
> I am personally very interested in the state of the
> tcp-window-tracking patch, especially for 2.6 kernels. It is very nice
> to know it is being ported for 2.6.
>
> What I would be doing would either be a patch that depends on the
> tcp-window-tracking patch to work at all - or it would do something
> similar itself which would most probably make it conflict with the
> tcp-window-tracking patch.
>
> Does anyone know of the status it? How far is it from being integrated
> into the mainline netfilter? Are there any blocking issues why it
> cannot be or will not be integrated?

Weren't it be ported to 2.6, couldn't it been submitted into
inclusion :-).

Thanks to Willy Tarreau's suggestions, patches and heavy testings, the
patch matured a lot. I'd say there is nothing against submitting it into
the kernel in the near future.

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

Re: tcp-window-tracking for 2.6? (Re: [PATCH] patch-o-matic-ng reorganization and updated patches)

[Herve Eychenne]

On Fri, Feb 20, 2004 at 10:12:24AM +0100, Jozsef Kadlecsik wrote:

> On Thu, 19 Feb 2004, Nuutti Kotivuori wrote:

> > > - 2.6 ports of tcp-window-tracking and conntrack_foo patches
> >
> > I am personally very interested in the state of the
> > tcp-window-tracking patch, especially for 2.6 kernels. It is very nice
> > to know it is being ported for 2.6.
> >
> > What I would be doing would either be a patch that depends on the
> > tcp-window-tracking patch to work at all - or it would do something
> > similar itself which would most probably make it conflict with the
> > tcp-window-tracking patch.
> >
> > Does anyone know of the status it? How far is it from being integrated
> > into the mainline netfilter? Are there any blocking issues why it
> > cannot be or will not be integrated?

> Weren't it be ported to 2.6, couldn't it been submitted into
> inclusion :-).

> Thanks to Willy Tarreau's suggestions, patches and heavy testings, the
> patch matured a lot. I'd say there is nothing against submitting it into
> the kernel in the near future.

At the workshop, you presented benchmarks showing that the box seemed
to run faster with tcp-window-tracking patch (!), surprising us all very
much. In the meanwhile, could someone figure out why?

 Herve

-- 
 _
(°=  Hervé Eychenne
//)
v_/_ WallFire project:  http://www.wallfire.org/

Re: tcp-window-tracking for 2.6? (Re: [PATCH] patch-o-matic-ng reorganization and updated patches)

[Jozsef Kadlecsik]

On Fri, 20 Feb 2004, Herve Eychenne wrote:

> At the workshop, you presented benchmarks showing that the box seemed
> to run faster with tcp-window-tracking patch (!), surprising us all very
> much. In the meanwhile, could someone figure out why?

No :-(.

I'm going to repeat the tests in a more real environment (10 servers and
40 clients). As usual there are time constraints, because after the
beginning of March the machines won't be available for me for testing.

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

Re: tcp-window-tracking for 2.6? (Re: [PATCH] patch-o-matic-ng reorganization and updated patches)

[Andrew Hall]

Jozsef,

There seems to be a dependancy on nf_log for the tcp-windowing patch? Using
the latest CVS patch-o-matic and 2.6, the nf_log patch seems to fail ( says
it's missing one needed file). Is this intentional or should we expect the
2.6 nf_log+tcp-windowing patches to install cleanly now?

Thanks,
Andrew.
----- Original Message -----
From: "Jozsef Kadlecsik" <kadlec@blackhole.kfki.hu>
To: "Nuutti Kotivuori" <naked@iki.fi>
Cc: <netfilter-devel@lists.netfilter.org>
Sent: Friday, February 20, 2004 8:12 PM
Subject: Re: tcp-window-tracking for 2.6? (Re: [PATCH] patch-o-matic-ng
reorganization and updated patches)


> On Thu, 19 Feb 2004, Nuutti Kotivuori wrote:
>
> > > - 2.6 ports of tcp-window-tracking and conntrack_foo patches
> >
> > I am personally very interested in the state of the
> > tcp-window-tracking patch, especially for 2.6 kernels. It is very nice
> > to know it is being ported for 2.6.
> >
> > What I would be doing would either be a patch that depends on the
> > tcp-window-tracking patch to work at all - or it would do something
> > similar itself which would most probably make it conflict with the
> > tcp-window-tracking patch.
> >
> > Does anyone know of the status it? How far is it from being integrated
> > into the mainline netfilter? Are there any blocking issues why it
> > cannot be or will not be integrated?
>
> Weren't it be ported to 2.6, couldn't it been submitted into
> inclusion :-).
>
> Thanks to Willy Tarreau's suggestions, patches and heavy testings, the
> patch matured a lot. I'd say there is nothing against submitting it into
> the kernel in the near future.
>
> 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
>
>

Re: tcp-window-tracking for 2.6? (Re: [PATCH] patch-o-matic-ng reorganization and updated patches)

[Jozsef Kadlecsik]

On Fri, 20 Feb 2004, Andrew Hall wrote:

> There seems to be a dependancy on nf_log for the tcp-windowing patch? Using
> the latest CVS patch-o-matic and 2.6, the nf_log patch seems to fail ( says
> it's missing one needed file). Is this intentional or should we expect the
> 2.6 nf_log+tcp-windowing patches to install cleanly now?

The logging part from the TCP window tracking patch was ripped off,
because there is now a commong logging ifrastructure (nf-log). Thus the
dependency is unavoidable.

The 2.6 nf-log patch fails from pom-ng? Then I have to double check it
again. Thank you for notifying about the problem.

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