mpls-linux-general — Mailing list for general discussion about MPLS for Linux

[mpls-linux-general] [mpls-general] problem with iproute2 setup

[vittorio_manetti <vit...@li...>]

Hi, 
I want to install iproute2 for linux-mpls (kernel 2.6.9). 
I have =
done two attempts. First time I have tried to install iproute2-2.6.9-ss04=
0831 later have applied the patch 
iproute2.diff. This is the error mess=
age : 
 
#make[1]: Entering directory `/iproute2/misc' 
#gcc -D_GNU_SO=
URCE -O2 -Wstrict-prototypes -Wall -g -I../include -DRESOLVE_HOSTNAMES   =
-c -o ss.o ss.c 
#In file included from /usr/include/asm/byteorder.h:57,=
 
#                 from ss.c:36: 
#/usr/include/linux/byteorder/little=
_endian.h:43: parse error before "__cpu_to_le64p" 
#/usr/include/linux/b=
yteorder/little_endian.h: In function `__cpu_to_le64p': 
#/usr/include/l=
inux/byteorder/little_endian.h:45: `__le64' undeclared (first use in this=
 function) 
#/usr/include/linux/byteorder/little_endian.h:45: (Each unde=
clared identifier is reported only once 
#/usr/include/linux/byteorder/l=
ittle_endian.h:45: for each function it appears in.) 
#/usr/include/linu=
x/byteorder/little_endian.h: At top level: 
#/usr/include/linux/byteorde=
r/little_endian.h:47: parse error before '*' token 
#/usr/include/linux/=
byteorder/little_endian.h: In function `__le64_to_cpup': 
#/usr/include/=
linux/byteorder/little_endian.h:49: `p' undeclared (first use in this fun=
ction) 
#/usr/include/linux/byteorder/little_endian.h: At top level: =0D
=
#/usr/include/linux/byteorder/little_endian.h:51: parse error before "__c=
pu_to_le32p" 
#/usr/include/linux/byteorder/little_endian.h: In function=
 `__cpu_to_le32p': 
#/usr/include/linux/byteorder/little_endian.h:53: `_=
_le32' undeclared (first use in this function) 
#/usr/include/linux/byte=
order/little_endian.h: At top level: 
#/usr/include/linux/byteorder/litt=
le_endian.h:55: parse error before '*' token 
#/usr/include/linux/byteor=
der/little_endian.h: In function `__le32_to_cpup': 
#/usr/include/linux/=
byteorder/little_endian.h:57: `p' undeclared (first use in this function)=
 
#/usr/include/linux/byteorder/little_endian.h: At top level: 
#/usr/i=
nclude/linux/byteorder/little_endian.h:59: parse error before "__cpu_to_l=
e16p" 
#/usr/include/linux/byteorder/little_endian.h: In function `__cpu=
_to_le16p': 
#/usr/include/linux/byteorder/little_endian.h:61: `__le16' =
undeclared (first use in this function) 
#/usr/include/linux/byteorder/l=
ittle_endian.h: At top level: 
#/usr/include/linux/byteorder/little_endi=
an.h:63: parse error before '*' token 
#/usr/include/linux/byteorder/lit=
tle_endian.h: In function `__le16_to_cpup': 
#/usr/include/linux/byteord=
er/little_endian.h:65: `p' undeclared (first use in this function) 
#/us=
r/include/linux/byteorder/little_endian.h: At top level: 
#/usr/include/=
linux/byteorder/little_endian.h:67: parse error before "__cpu_to_be64p" =0D
=
#/usr/include/linux/byteorder/little_endian.h: In function `__cpu_to_be64=
p': 
#/usr/include/linux/byteorder/little_endian.h:69: `__be64' undeclar=
ed (first use in this function) 
#/usr/include/linux/byteorder/little_en=
dian.h:69: parse error before "__swab64p" 
#/usr/include/linux/byteorder=
/little_endian.h: At top level: 
#/usr/include/linux/byteorder/little_en=
dian.h:71: parse error before '*' token 
#/usr/include/linux/byteorder/l=
ittle_endian.h: In function `__be64_to_cpup': 
#/usr/include/linux/byteo=
rder/little_endian.h:73: `p' undeclared (first use in this function) 
#/=
usr/include/linux/byteorder/little_endian.h: At top level: 
#/usr/includ=
e/linux/byteorder/little_endian.h:75: parse error before "__cpu_to_be32p"=
 
#/usr/include/linux/byteorder/little_endian.h: In function `__cpu_to_b=
e32p': 
#/usr/include/linux/byteorder/little_endian.h:77: `__be32' undec=
lared (first use in this function) 
#/usr/include/linux/byteorder/little=
_endian.h:77: parse error before "__swab32p" 
#/usr/include/linux/byteor=
der/little_endian.h: At top level: 
#/usr/include/linux/byteorder/little=
_endian.h:79: parse error before '*' token 
#/usr/include/linux/byteorde=
r/little_endian.h: In function `__be32_to_cpup': 
#/usr/include/linux/by=
teorder/little_endian.h:81: `p' undeclared (first use in this function) =0D
=
#/usr/include/linux/byteorder/little_endian.h: At top level: 
#/usr/incl=
ude/linux/byteorder/little_endian.h:83: parse error before "__cpu_to_be16=
p" 
#/usr/include/linux/byteorder/little_endian.h: In function `__cpu_to=
_be16p': 
#/usr/include/linux/byteorder/little_endian.h:85: `__be16' und=
eclared (first use in this function) 
#/usr/include/linux/byteorder/litt=
le_endian.h:85: parse error before "__swab16p" 
#/usr/include/linux/byte=
order/little_endian.h: At top level: 
#/usr/include/linux/byteorder/litt=
le_endian.h:87: parse error before '*' token 
#/usr/include/linux/byteor=
der/little_endian.h: In function `__be16_to_cpup': 
#/usr/include/linux/=
byteorder/little_endian.h:89: `p' undeclared (first use in this function)=
 
#In file included from ss.c:37: 
#/usr/include/linux/tcp.h: At top le=
vel: 
#/usr/include/linux/tcp.h:105: `__be32' undeclared here (not in a =
function) 
#/usr/include/linux/tcp.h:105: enumerator value for `TCP_FLAG=
_CWR' not integer constant 
#/usr/include/linux/tcp.h:106: `__be32' unde=
clared here (not in a function) 
#/usr/include/linux/tcp.h:106: enumerat=
or value for `TCP_FLAG_ECE' not integer constant 
#/usr/include/linux/tc=
p.h:107: `__be32' undeclared here (not in a function) 
#/usr/include/lin=
ux/tcp.h:107: enumerator value for `TCP_FLAG_URG' not integer constant =0D
=
#/usr/include/linux/tcp.h:108: `__be32' undeclared here (not in a functio=
n) 
#/usr/include/linux/tcp.h:108: enumerator value for `TCP_FLAG_ACK' n=
ot integer constant 
#/usr/include/linux/tcp.h:109: `__be32' undeclared =
here (not in a function) 
#/usr/include/linux/tcp.h:109: enumerator valu=
e for `TCP_FLAG_PSH' not integer constant 
#/usr/include/linux/tcp.h:110=
: `__be32' undeclared here (not in a function) 
#/usr/include/linux/tcp.=
h:110: enumerator value for `TCP_FLAG_RST' not integer constant 
#/usr/i=
nclude/linux/tcp.h:111: `__be32' undeclared here (not in a function) 
#/=
usr/include/linux/tcp.h:111: enumerator value for `TCP_FLAG_SYN' not inte=
ger constant 
#/usr/include/linux/tcp.h:112: `__be32' undeclared here (n=
ot in a function) 
#/usr/include/linux/tcp.h:112: enumerator value for `=
TCP_FLAG_FIN' not integer constant 
#/usr/include/linux/tcp.h:113: `__be=
32' undeclared here (not in a function) 
#/usr/include/linux/tcp.h:113: =
enumerator value for `TCP_RESERVED_BITS' not integer constant 
#/usr/inc=
lude/linux/tcp.h:114: `__be32' undeclared here (not in a function) 
#/us=
r/include/linux/tcp.h:115: enumerator value for `TCP_DATA_OFFSET' not int=
eger constant 
#make[1]: *** [ss.o] Error 1 
#make[1]: Leaving director=
y `/iproute2/misc' 
#make: *** [all] Error 2 
 
I have subsequently tr=
ied to install the package iproute-2.6.9-3mpls1_946.i386.rpm, but the fol=
lowing error 
message appears : 
 
#file /etc/iproute2/rt_realms from =
install of iproute-2.6.9-3mpls1_946 conflicts with file from package 
ip=
route2-2.4.7-7mdk 
#file /etc/iproute2/rt_tables from install of iproute=
-2.6.9-3mpls1_946 conflicts with file from package 
iproute2-2.4.7-7mdk =

#file /sbin/ifcfg from install of iproute-2.6.9-3mpls1_946 conflicts wi=
th file from package iproute2-2.4.7-7mdk 
#file /sbin/ip from install of=
 iproute-2.6.9-3mpls1_946 conflicts with file from package iproute2-2.4.7=
-7mdk 
#file /sbin/rtmon from install of iproute-2.6.9-3mpls1_946 confli=
cts with file from package iproute2-2.4.7-7mdk 
#file /sbin/tc from inst=
all of iproute-2.6.9-3mpls1_946 conflicts with file from package iproute2=
-2.4.7-7mdk 
 
I have not been able to uninstall the package iproute2-2=
.4.7-7mdk. Can someone help me? 
 
Another question: 
I have correctly=
 installed quagga-mpls and I have tried to use the commands show in the d=
ocument 
README.quagga; for every command the message always appears is =
: 
 
command not found 
 
Is the problem employed at the wrong versio=
n of iproute (2.4.7) ? 
 =0A=0A=0A=0A___________________________________=
_________________________=0A6X velocizzare la tua navigazione a 56k? 6X W=
eb Accelerator di Libero!=0AScaricalo su INTERNET GRATIS 6X http://www.li=
bero.it=0A

Re: [mpls-linux-general] bug with mpls tunnel ?

[James R. L. <jl...@mi...>]

Try using a dev name of mplsX:

   mpls tunnel add dev mpls0 ...

On Mon, May 23, 2005 at 11:18:46AM +0200, Bertrand Lemaitre wrote:
> Hi,
>=20
> I tested mpls tunnels in shell with "mpls tunnel add dev eth1 ..." but I =
already
> had the message RTNETLINK answers: Cannot allocate memory
>=20
>=20
> So, I try to make tunnel with zebra and I succeeded ...
>=20
> So, I think there is a problem with tunnels in shell
>=20
> Bertrand Lemaitre
>=20
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=3D7412&alloc_id=3D16344&op=3Dclick
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--=20
James R. Leu
jl...@mi...

Re: [mpls-linux-general] Which distribution ?

[James R. L. <jl...@mi...>]

I build all of my RPMS on Fedora Core 3.

On Mon, May 23, 2005 at 09:37:11AM +0200, Bertrand Lemaitre wrote:
> Hello,
>=20
> I tested MPLS / LINUX on a LFS. But, there are a lot of problems.
> So, I would like to test it with a dedicated distribution.
> James, can you tell me which distribution you used to compile your RPM ?
>=20
> Thanks and have a good day,
>=20
> Bertrand Lemaitre
>=20
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=3D7412&alloc_id=3D16344&op=3Dclick
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--=20
James R. Leu
jl...@mi...

[mpls-linux-general] bug with mpls tunnel ?

[Bertrand L. <ble...@in...>]

Hi,

I tested mpls tunnels in shell with "mpls tunnel add dev eth1 ..." but I already
had the message RTNETLINK answers: Cannot allocate memory


So, I try to make tunnel with zebra and I succeeded ...

So, I think there is a problem with tunnels in shell

Bertrand Lemaitre

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

[mpls-linux-general] Which distribution ?

[Bertrand L. <ble...@in...>]

Hello,

I tested MPLS / LINUX on a LFS. But, there are a lot of problems.
So, I would like to test it with a dedicated distribution.
James, can you tell me which distribution you used to compile your RPM ?

Thanks and have a good day,

Bertrand Lemaitre

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

[mpls-linux-general] MPLS Linux + linux-vrf

[A f. S. M. <afo...@ya...>]

Hi,
 
I've got the same problem than Irfan Azher (mailing list linux-vrf 2004-10-26), ie I would like to use linux-vrf with mpls-linux.
You told him it was not working together at the time. I would like to know if there have been some improvements.
 
Thanks
AFS MPLS

		
---------------------------------
 Découvrez le nouveau Yahoo! Mail : 1 Go d'espace de stockage pour vos mails, photos et vidéos !
Créez votre Yahoo! Mail  

Re: [mpls-linux-general] Use of set-exp

[James R. L. <jl...@mi...>]

exp2tc allows you to set the tcindex on the packet based on the incoming
EXP bits.  The tcindex can be used by TC schedulers.  Read up on TC to learn
how to utilize the tcindex.

On Thu, May 19, 2005 at 10:59:58PM +0200, Julien Cayssol wrote:
>=20
> yes but in a LSR how can setup priority between a paquet exp 1 and 6 ?
>=20
> James R. Leu wrote:
>=20
> >mpls ilm add label gen 16 labelspace 0 instructions pop exp2tc 0x0 0x10=
=20
> >0x1 0x11 0x2 0x12 0x3 0x13 0x4 0x14 0x5 0x15 0x6 0x16 0x7 0x17 peek
> >
> >(again I don't have access to my test machines so the above command is=
=20
> >untested)
> >
> >The above example maps each EXP value to a TC value which is 0x10 more.
> >You can choose to only map certain EXP and you can choose whatever TC=20
> >values
> >you want.
> >
> >BTW the 'mpls' command does have 'usage' output if you just type 'mpls'
> >by itself.  If the output is difficult to understand, please suggest=20
> >changes.
> >
> >On Thu, May 19, 2005 at 12:58:33PM +0200, Julien Cayssol wrote:
> >=20
> >
> >>Julien Cayssol wrote:
> >>
> >>  =20
> >>
> >>>James R. Leu wrote:
> >>>
> >>>    =20
> >>>
> >>>>On Wed, May 18, 2005 at 03:55:42PM +0200, jul...@aq... wrote:
> >>>>
> >>>>
> >>>>      =20
> >>>>
> >>>>>Hello,
> >>>>>
> >>>>>The search function on sourceforge for the mailing-list doesn't=20
> >>>>>works so i post
> >>>>>my question.
> >>>>>
> >>>>>With the command "mpls" you can specify the exp field.
> >>>>>
> >>>>>I don't understand what you must specify to make it work ?
> >>>>>does any body have an example ?
> >>>>>
> >>>>>        =20
> >>>>>
> >>>>It must be before the push instruction, for example:
> >>>>
> >>>> mpls nhlfe add key 0 instructions set-exp 0x2 push gen 16 nexthop=20
> >>>>eth0 ipv4 1.1.1.1
> >>>>
> >>>>I don't have access to my test boxes right now, so I can't confirm the
> >>>>actual command above works.
> >>>>
> >>>>What commands have you tried?  What is the resulting MPLS kernel debug
> >>>>that you get?
> >>>>
> >>>>
> >>>>      =20
> >>>>
> >>>i've try somes commandes but only with a "change key", the exp field=
=20
> >>>is use for classe services,
> >>>
> >>>Is it possible to map it with TC ?
> >>>
> >>>were can we use exp next ?
> >>>    =20
> >>>
> >>Helo,
> >>
> >>With iptable we set an exp to  a paquet mpls, on the nexthop an exp a=
=20
> >>6(dm) is faster than a exp 0(best effort).
> >>but when we simulate traffic there are no differents.
> >>
> >>We must use tc ? can i have somes informations, were to search, on the=
=20
> >>mailing liste a don't find a lot of informations.
> >>
> >>thanks.
> >>
> >>  =20
> >>
> >>>>      =20
> >>>>
> >>>>>Thanks a lot.
> >>>>>
> >>>>>
> >>>>>--=20
> >>>>>Julien CAYSSOL
> >>>>>INSA de LYON                  -=3D-      D?partement Telecom Service=
s=20
> >>>>>et Usages
> >>>>>http://www.aqwz.com/                                    T?l : +33=20
> >>>>>664 973 067
> >>>>>
> >>>>>
> >>>>>-------------------------------------------------------
> >>>>>This SF.Net email is sponsored by Oracle Space Sweepstakes
> >>>>>Want to be the first software developer in space?
> >>>>>Enter now for the Oracle Space Sweepstakes!
> >>>>>http://ads.osdn.com/?ad_id=3D7412&alloc_id=3D16344&op=3Dclick
> >>>>>_______________________________________________
> >>>>>mpls-linux-general mailing list
> >>>>>mpl...@li...
> >>>>>https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
> >>>>>
> >>>>>        =20
> >>>>>
> >>>>
> >>>>
> >>>>      =20
> >>>>
> >>>    =20
> >>>
> >>--=20
> >>---
> >>
> >>Julien Cayssol
> >>http://www.aqwz.com
> >>Mobile : 33 664 973 067
> >>  =20
> >>
> >
> >=20
> >
>=20
>=20
> --=20
> ---
>=20
> Julien Cayssol
> http://www.aqwz.com
> Mobile : 33 664 973 067

--=20
James R. Leu
jl...@mi...

Re: [mpls-linux-general] Use of set-exp

[Julien C. <jul...@aq...>]

yes but in a LSR how can setup priority between a paquet exp 1 and 6 ?

James R. Leu wrote:

>mpls ilm add label gen 16 labelspace 0 instructions pop exp2tc 0x0 0x10 0x1 0x11 0x2 0x12 0x3 0x13 0x4 0x14 0x5 0x15 0x6 0x16 0x7 0x17 peek
>
>(again I don't have access to my test machines so the above command is untested)
>
>The above example maps each EXP value to a TC value which is 0x10 more.
>You can choose to only map certain EXP and you can choose whatever TC values
>you want.
>
>BTW the 'mpls' command does have 'usage' output if you just type 'mpls'
>by itself.  If the output is difficult to understand, please suggest changes.
>
>On Thu, May 19, 2005 at 12:58:33PM +0200, Julien Cayssol wrote:
>  
>
>>Julien Cayssol wrote:
>>
>>    
>>
>>>James R. Leu wrote:
>>>
>>>      
>>>
>>>>On Wed, May 18, 2005 at 03:55:42PM +0200, jul...@aq... wrote:
>>>>
>>>>
>>>>        
>>>>
>>>>>Hello,
>>>>>
>>>>>The search function on sourceforge for the mailing-list doesn't 
>>>>>works so i post
>>>>>my question.
>>>>>
>>>>>With the command "mpls" you can specify the exp field.
>>>>>
>>>>>I don't understand what you must specify to make it work ?
>>>>>does any body have an example ?
>>>>> 
>>>>>          
>>>>>
>>>>It must be before the push instruction, for example:
>>>>
>>>>  mpls nhlfe add key 0 instructions set-exp 0x2 push gen 16 nexthop 
>>>>eth0 ipv4 1.1.1.1
>>>>
>>>>I don't have access to my test boxes right now, so I can't confirm the
>>>>actual command above works.
>>>>
>>>>What commands have you tried?  What is the resulting MPLS kernel debug
>>>>that you get?
>>>>
>>>>
>>>>        
>>>>
>>>i've try somes commandes but only with a "change key", the exp field 
>>>is use for classe services,
>>>
>>>Is it possible to map it with TC ?
>>>
>>>were can we use exp next ?
>>>      
>>>
>>Helo,
>>
>>With iptable we set an exp to  a paquet mpls, on the nexthop an exp a 
>>6(dm) is faster than a exp 0(best effort).
>>but when we simulate traffic there are no differents.
>>
>>We must use tc ? can i have somes informations, were to search, on the 
>>mailing liste a don't find a lot of informations.
>>
>>thanks.
>>
>>    
>>
>>>>        
>>>>
>>>>>Thanks a lot.
>>>>>
>>>>>
>>>>>-- 
>>>>>Julien CAYSSOL
>>>>>INSA de LYON                  -=-      D?partement Telecom Services 
>>>>>et Usages
>>>>>http://www.aqwz.com/                                    T?l : +33 
>>>>>664 973 067
>>>>>
>>>>>
>>>>>-------------------------------------------------------
>>>>>This SF.Net email is sponsored by Oracle Space Sweepstakes
>>>>>Want to be the first software developer in space?
>>>>>Enter now for the Oracle Space Sweepstakes!
>>>>>http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
>>>>>_______________________________________________
>>>>>mpls-linux-general mailing list
>>>>>mpl...@li...
>>>>>https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
>>>>> 
>>>>>          
>>>>>
>>>>
>>>>
>>>>        
>>>>
>>>      
>>>
>>-- 
>>---
>>
>>Julien Cayssol
>>http://www.aqwz.com
>>Mobile : 33 664 973 067
>>    
>>
>
>  
>


-- 
---

Julien Cayssol
http://www.aqwz.com
Mobile : 33 664 973 067

Re: [mpls-linux-general] Use of set-exp

[James R. L. <jl...@mi...>]

mpls ilm add label gen 16 labelspace 0 instructions pop exp2tc 0x0 0x10 0x1=
 0x11 0x2 0x12 0x3 0x13 0x4 0x14 0x5 0x15 0x6 0x16 0x7 0x17 peek

(again I don't have access to my test machines so the above command is unte=
sted)

The above example maps each EXP value to a TC value which is 0x10 more.
You can choose to only map certain EXP and you can choose whatever TC values
you want.

BTW the 'mpls' command does have 'usage' output if you just type 'mpls'
by itself.  If the output is difficult to understand, please suggest change=
s.

On Thu, May 19, 2005 at 12:58:33PM +0200, Julien Cayssol wrote:
> Julien Cayssol wrote:
>=20
> >James R. Leu wrote:
> >
> >>On Wed, May 18, 2005 at 03:55:42PM +0200, jul...@aq... wrote:
> >>=20
> >>
> >>>Hello,
> >>>
> >>>The search function on sourceforge for the mailing-list doesn't=20
> >>>works so i post
> >>>my question.
> >>>
> >>>With the command "mpls" you can specify the exp field.
> >>>
> >>>I don't understand what you must specify to make it work ?
> >>>does any body have an example ?
> >>> =20
> >>
> >>
> >>It must be before the push instruction, for example:
> >>
> >>   mpls nhlfe add key 0 instructions set-exp 0x2 push gen 16 nexthop=20
> >>eth0 ipv4 1.1.1.1
> >>
> >>I don't have access to my test boxes right now, so I can't confirm the
> >>actual command above works.
> >>
> >>What commands have you tried?  What is the resulting MPLS kernel debug
> >>that you get?
> >>=20
> >>
> >
> >i've try somes commandes but only with a "change key", the exp field=20
> >is use for classe services,
> >
> >Is it possible to map it with TC ?
> >
> >were can we use exp next ?
>=20
> Helo,
>=20
> With iptable we set an exp to  a paquet mpls, on the nexthop an exp a=20
> 6(dm) is faster than a exp 0(best effort).
> but when we simulate traffic there are no differents.
>=20
> We must use tc ? can i have somes informations, were to search, on the=20
> mailing liste a don't find a lot of informations.
>=20
> thanks.
>=20
> >>=20
> >>
> >>>Thanks a lot.
> >>>
> >>>
> >>>--=20
> >>>Julien CAYSSOL
> >>>INSA de LYON                  -=3D-      D?partement Telecom Services=
=20
> >>>et Usages
> >>>http://www.aqwz.com/                                    T?l : +33=20
> >>>664 973 067
> >>>
> >>>
> >>>-------------------------------------------------------
> >>>This SF.Net email is sponsored by Oracle Space Sweepstakes
> >>>Want to be the first software developer in space?
> >>>Enter now for the Oracle Space Sweepstakes!
> >>>http://ads.osdn.com/?ad_id=3D7412&alloc_id=3D16344&op=3Dclick
> >>>_______________________________________________
> >>>mpls-linux-general mailing list
> >>>mpl...@li...
> >>>https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
> >>> =20
> >>
> >>
> >>=20
> >>
> >
> >
>=20
>=20
> --=20
> ---
>=20
> Julien Cayssol
> http://www.aqwz.com
> Mobile : 33 664 973 067

--=20
James R. Leu
jl...@mi...

Re: [mpls-linux-general] Use of set-exp

[Julien C. <jul...@aq...>]

Julien Cayssol wrote:

> James R. Leu wrote:
>
>> On Wed, May 18, 2005 at 03:55:42PM +0200, jul...@aq... wrote:
>>  
>>
>>> Hello,
>>>
>>> The search function on sourceforge for the mailing-list doesn't 
>>> works so i post
>>> my question.
>>>
>>> With the command "mpls" you can specify the exp field.
>>>
>>> I don't understand what you must specify to make it work ?
>>> does any body have an example ?
>>>   
>>
>>
>> It must be before the push instruction, for example:
>>
>>    mpls nhlfe add key 0 instructions set-exp 0x2 push gen 16 nexthop 
>> eth0 ipv4 1.1.1.1
>>
>> I don't have access to my test boxes right now, so I can't confirm the
>> actual command above works.
>>
>> What commands have you tried?  What is the resulting MPLS kernel debug
>> that you get?
>>  
>>
>
> i've try somes commandes but only with a "change key", the exp field 
> is use for classe services,
>
> Is it possible to map it with TC ?
>
> were can we use exp next ?

Helo,

With iptable we set an exp to  a paquet mpls, on the nexthop an exp a 
6(dm) is faster than a exp 0(best effort).
but when we simulate traffic there are no differents.

We must use tc ? can i have somes informations, were to search, on the 
mailing liste a don't find a lot of informations.

thanks.

>>  
>>
>>> Thanks a lot.
>>>
>>>
>>> -- 
>>> Julien CAYSSOL
>>> INSA de LYON                  -=-      D?partement Telecom Services 
>>> et Usages
>>> http://www.aqwz.com/                                    T?l : +33 
>>> 664 973 067
>>>
>>>
>>> -------------------------------------------------------
>>> This SF.Net email is sponsored by Oracle Space Sweepstakes
>>> Want to be the first software developer in space?
>>> Enter now for the Oracle Space Sweepstakes!
>>> http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
>>> _______________________________________________
>>> mpls-linux-general mailing list
>>> mpl...@li...
>>> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
>>>   
>>
>>
>>  
>>
>
>


-- 
---

Julien Cayssol
http://www.aqwz.com
Mobile : 33 664 973 067

Re: [mpls-linux-general] Use of set-exp

[Julien C. <jul...@aq...>]

James R. Leu wrote:

>On Wed, May 18, 2005 at 03:55:42PM +0200, jul...@aq... wrote:
>  
>
>>Hello,
>>
>>The search function on sourceforge for the mailing-list doesn't works so i post
>>my question.
>>
>>With the command "mpls" you can specify the exp field.
>>
>>I don't understand what you must specify to make it work ?
>>does any body have an example ?
>>    
>>
>
>It must be before the push instruction, for example:
>
>    mpls nhlfe add key 0 instructions set-exp 0x2 push gen 16 nexthop eth0 ipv4 1.1.1.1
>
>I don't have access to my test boxes right now, so I can't confirm the
>actual command above works.
>
>What commands have you tried?  What is the resulting MPLS kernel debug
>that you get?
>  
>

i've try somes commandes but only with a "change key", the exp field is 
use for classe services,

Is it possible to map it with TC ?

were can we use exp next ?

>  
>
>>Thanks a lot.
>>
>>
>>-- 
>>Julien CAYSSOL
>>INSA de LYON                  -=-      D?partement Telecom Services et Usages
>>http://www.aqwz.com/                                    T?l : +33 664 973 067
>>
>>
>>-------------------------------------------------------
>>This SF.Net email is sponsored by Oracle Space Sweepstakes
>>Want to be the first software developer in space?
>>Enter now for the Oracle Space Sweepstakes!
>>http://ads.osdn.com/?ad_id=7412&alloc_id=16344&op=click
>>_______________________________________________
>>mpls-linux-general mailing list
>>mpl...@li...
>>https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
>>    
>>
>
>  
>


-- 
---

Julien Cayssol
http://www.aqwz.com
Mobile : 33 664 973 067

Re: [mpls-linux-general] Use of set-exp

[James R. L. <jl...@mi...>]

On Wed, May 18, 2005 at 03:55:42PM +0200, jul...@aq... wrote:
> Hello,
>=20
> The search function on sourceforge for the mailing-list doesn't works so =
i post
> my question.
>=20
> With the command "mpls" you can specify the exp field.
>=20
> I don't understand what you must specify to make it work ?
> does any body have an example ?

It must be before the push instruction, for example:

    mpls nhlfe add key 0 instructions set-exp 0x2 push gen 16 nexthop eth0 =
ipv4 1.1.1.1

I don't have access to my test boxes right now, so I can't confirm the
actual command above works.

What commands have you tried?  What is the resulting MPLS kernel debug
that you get?

>=20
> Thanks a lot.
>=20
>=20
> --=20
> Julien CAYSSOL
> INSA de LYON                  -=3D-      D?partement Telecom Services et =
Usages
> http://www.aqwz.com/                                    T?l : +33 664 973=
 067
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=3D7412&alloc_id=3D16344&op=3Dclick
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--=20
James R. Leu
jl...@mi...

[mpls-linux-general] Use of set-exp

[<jul...@aq...>]

Hello,

The search function on sourceforge for the mailing-list doesn't works so i post
my question.

With the command "mpls" you can specify the exp field.

I don't understand what you must specify to make it work ?
does any body have an example ?

Thanks a lot.


-- 
Julien CAYSSOL
INSA de LYON                  -=-      Département Telecom Services et Usages
http://www.aqwz.com/                                    Tèl : +33 664 973 067

Re: [mpls-linux-general] Dubts

[James R. L. <jl...@mi...>]

Hello Miquel,

A quick search of the mpls-linux-general mailing list will point
out that ldpd is not ready for users.  If you are willing to dig into the
code and help fix bugs then get the most recent version of quagga-mpls and
ldp-portable from my development tree.  See http://mpls-linux.sf.net/
for more details.  Otherwise I would suggest holding of on trying to
use ldpd.  Static LSPs configure via the zebra CLI should work.

On Tue, May 17, 2005 at 06:38:12PM +0200, Miquel Casanovas Puigvi wrote:
> Hello, I have any dubts in protocol LDP, the first is:
>=20
> LDP_2# show ldp data
>   192.168.0.0/24  local binding:   label: gen 10000
>   192.168.3.0/24  local binding:   label: gen 10001
>   192.168.0.0/24  remote binding:  no outlabel lsr: 192.168.0.1:0
>   192.168.1.0/24  remote binding:  no outlabel lsr: 192.168.0.1:0
>   192.168.2.0/24  remote binding:  no outlabel lsr: 192.168.3.3:0
>   192.168.0.0/24  local binding:   label: gen 10002
>   192.168.3.0/24  local binding:   label: gen 10003
>   192.168.3.0/24  remote binding:  no outlabel lsr: 192.168.3.3:0
>   192.168.4.0/24  remote binding:  no outlabel lsr: 192.168.3.3:0
>=20
> But, I assign the remote binding manually??
>=20
> The second is , when run the ldpd:
>=20
> [root@uni ldp-portable]# ldpd -P 1026
> 2005/05/09 18:38:46 LDP: old umask 23 127
> 2005/05/09 18:38:46 LDP: interface add eth0 index 2 flags 4098 metric 1
> mtu 1500
> 2005/05/09 18:38:46 LDP: interface add eth1 index 3 flags 4098 metric 1
> mtu 1500
> 2005/05/09 18:38:46 LDP: interface add eth2 index 4 flags 4098 metric 1
> mtu 1500
> 2005/05/09 18:38:46 LDP: interface add eth3 index 5 flags 4163 metric 1
> mtu 1500
> 2005/05/09 18:38:46 LDP: address add 192.168.3.2 to interface eth3
> 2005/05/09 18:38:46 LDP: interface add eth4 index 6 flags 4163 metric 1
> mtu 1500
> 2005/05/09 18:38:46 LDP: address add 192.168.0.2 to interface eth4
> 2005/05/09 18:38:46 LDP: interface add lo index 1 flags 73 metric 1 mtu 1=
6436
> 2005/05/09 18:38:46 LDP: address add 127.0.0.1 to interface lo
> 2005/05/09 18:38:46 LDP: interface add sit0 index 7 flags 128 metric 1 mtu
> 1480
> 2005/05/09 18:38:46 LDP: router-id change 192.168.0.2
> 2005/05/09 18:38:46 LDP: route 192.168.0.0/24
> 2005/05/09 18:38:46 LDP:        attached
> 2005/05/09 18:38:46 LDP:        num nexthop 1
> 2005/05/09 18:38:46 LDP:        nexthop 1.0.0.0
> 2005/05/09 18:38:46 LDP:        add
> 2005/05/09 18:38:46 LDP: route 192.168.3.0/24
> 2005/05/09 18:38:46 LDP:        attached
> 2005/05/09 18:38:46 LDP:        num nexthop 1
> 2005/05/09 18:38:46 LDP:        nexthop 1.0.0.0
> 2005/05/09 18:38:46 LDP:        add
> 2005/05/09 18:38:46 LDP: route 192.168.1.0/24
> 2005/05/09 18:38:46 LDP:        num nexthop 1
> 2005/05/09 18:38:46 LDP:        nexthop 5.192.168.0
> 2005/05/09 18:38:46 LDP:        add
> 2005/05/09 18:38:46 LDP: route 192.168.2.0/24
> 2005/05/09 18:38:46 LDP:        num nexthop 1
> 2005/05/09 18:38:46 LDP:        nexthop 5.192.168.3
> 2005/05/09 18:38:46 LDP:        add
> 2005/05/09 18:38:46 LDP: route 192.168.4.0/24
> 2005/05/09 18:38:46 LDP:        num nexthop 1
> 2005/05/09 18:38:46 LDP:        nexthop 5.192.168.3
> 2005/05/09 18:38:46 LDP:        add
> OUT: ldp_session_attempt_setup: MPLS_NON_BLOCKING
>=20
>=20
> What is nexthop ??? t'he 1.0.0.0 isn't ip of my LAN, 5.192.168.3 isn't to=
o.
> What is??
>=20
> And the last, when execute the ripd deamon, I see the follow line:
>=20
> [root@uni quagga]# ripd -d
> 2005/05/09 18:24:51 RIP: can't setsockopt SO_RCVBUF
>=20
> The deamon run correctly, but I know the why?
>=20
> thank you!!!
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=3D7412&alloc_id=3D16344&op=3Dclick
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--=20
James R. Leu
jl...@mi...

[mpls-linux-general] Dubts

[Miquel C. P. <u10...@co...>]

Hello, I have any dubts in protocol LDP, the first is:

LDP_2# show ldp data
  192.168.0.0/24  local binding:   label: gen 10000
  192.168.3.0/24  local binding:   label: gen 10001
  192.168.0.0/24  remote binding:  no outlabel lsr: 192.168.0.1:0
  192.168.1.0/24  remote binding:  no outlabel lsr: 192.168.0.1:0
  192.168.2.0/24  remote binding:  no outlabel lsr: 192.168.3.3:0
  192.168.0.0/24  local binding:   label: gen 10002
  192.168.3.0/24  local binding:   label: gen 10003
  192.168.3.0/24  remote binding:  no outlabel lsr: 192.168.3.3:0
  192.168.4.0/24  remote binding:  no outlabel lsr: 192.168.3.3:0

But, I assign the remote binding manually??

The second is , when run the ldpd:

[root@uni ldp-portable]# ldpd -P 1026
2005/05/09 18:38:46 LDP: old umask 23 127
2005/05/09 18:38:46 LDP: interface add eth0 index 2 flags 4098 metric 1
mtu 1500
2005/05/09 18:38:46 LDP: interface add eth1 index 3 flags 4098 metric 1
mtu 1500
2005/05/09 18:38:46 LDP: interface add eth2 index 4 flags 4098 metric 1
mtu 1500
2005/05/09 18:38:46 LDP: interface add eth3 index 5 flags 4163 metric 1
mtu 1500
2005/05/09 18:38:46 LDP: address add 192.168.3.2 to interface eth3
2005/05/09 18:38:46 LDP: interface add eth4 index 6 flags 4163 metric 1
mtu 1500
2005/05/09 18:38:46 LDP: address add 192.168.0.2 to interface eth4
2005/05/09 18:38:46 LDP: interface add lo index 1 flags 73 metric 1 mtu 16436
2005/05/09 18:38:46 LDP: address add 127.0.0.1 to interface lo
2005/05/09 18:38:46 LDP: interface add sit0 index 7 flags 128 metric 1 mtu
1480
2005/05/09 18:38:46 LDP: router-id change 192.168.0.2
2005/05/09 18:38:46 LDP: route 192.168.0.0/24
2005/05/09 18:38:46 LDP:        attached
2005/05/09 18:38:46 LDP:        num nexthop 1
2005/05/09 18:38:46 LDP:        nexthop 1.0.0.0
2005/05/09 18:38:46 LDP:        add
2005/05/09 18:38:46 LDP: route 192.168.3.0/24
2005/05/09 18:38:46 LDP:        attached
2005/05/09 18:38:46 LDP:        num nexthop 1
2005/05/09 18:38:46 LDP:        nexthop 1.0.0.0
2005/05/09 18:38:46 LDP:        add
2005/05/09 18:38:46 LDP: route 192.168.1.0/24
2005/05/09 18:38:46 LDP:        num nexthop 1
2005/05/09 18:38:46 LDP:        nexthop 5.192.168.0
2005/05/09 18:38:46 LDP:        add
2005/05/09 18:38:46 LDP: route 192.168.2.0/24
2005/05/09 18:38:46 LDP:        num nexthop 1
2005/05/09 18:38:46 LDP:        nexthop 5.192.168.3
2005/05/09 18:38:46 LDP:        add
2005/05/09 18:38:46 LDP: route 192.168.4.0/24
2005/05/09 18:38:46 LDP:        num nexthop 1
2005/05/09 18:38:46 LDP:        nexthop 5.192.168.3
2005/05/09 18:38:46 LDP:        add
OUT: ldp_session_attempt_setup: MPLS_NON_BLOCKING


What is nexthop ??? t'he 1.0.0.0 isn't ip of my LAN, 5.192.168.3 isn't too.
What is??

And the last, when execute the ripd deamon, I see the follow line:

[root@uni quagga]# ripd -d
2005/05/09 18:24:51 RIP: can't setsockopt SO_RCVBUF

The deamon run correctly, but I know the why?

thank you!!!

Re: [mpls-linux-general] clear tables

[James R. L. <jl...@mi...>]

On Fri, May 13, 2005 at 03:05:50PM +0200, A fond Sur MPLS wrote:
> Hi
> I would like to know if there is a command that allows
> destroying all the tables (NHLFE, ILM, XC), ie
> reseting the configuration.
>=20
> I've tried some commands such as mpls nhlfe del * but
> nothing seems to work...

Currently this functionally is not exposed to the user via the 'mpls'
command.  Although it wouldn't be hard to write a wrapper script that
uses the output of 'mpls ilm show' and issues all of the delete comamnds.

> Thanks for your help.
>=20
>=20
> =09
>=20
> =09
> 	=09
> _________________________________________________________________________=
____=20
> D?couvrez le nouveau Yahoo! Mail : 1 Go d'espace de stockage pour vos mai=
ls, photos et vid?os !=20
> Cr?ez votre Yahoo! Mail sur http://fr.mail.yahoo.com
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=3D7393&alloc_id=3D16281&op=3Dclick
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--=20
James R. Leu
jl...@mi...

Re: [mpls-linux-general] clear tables

[Vincent U. <vin...@im...>]

Le vendredi 13 mai 2005 =C3=A0 15:05 +0200, A fond Sur MPLS a =C3=A9crit :
> Hi
> I would like to know if there is a command that allows
> destroying all the tables (NHLFE, ILM, XC), ie
> reseting the configuration.
>=20
> I've tried some commands such as mpls nhlfe del * but
> nothing seems to work...

Something like this should work:

for i in $(mpls nhlfe show | grep NHLFE | cut -d ' ' -f 4); do
   mpls nhlfe del $i
done

ILM and XC are left as exercises ;-)

Vincent

--=20
Les gens heureux ne sont pas press=C3=A9s.

[mpls-linux-general] clear tables

[A f. S. M. <afo...@ya...>]

Hi
I would like to know if there is a command that allows
destroying all the tables (NHLFE, ILM, XC), ie
reseting the configuration.

I've tried some commands such as mpls nhlfe del * but
nothing seems to work...

Thanks for your help.


	

	
		
_____________________________________________________________________________ 
Découvrez le nouveau Yahoo! Mail : 1 Go d'espace de stockage pour vos mails, photos et vidéos ! 
Créez votre Yahoo! Mail sur http://fr.mail.yahoo.com

Re: [mpls-linux-general] Problems with linux 2.6.9 or linux 2.6.11.8

[James R. L. <jl...@mi...>]

Don't compile MPLS or the IPv4|6 MPLS hook as modules.

On Wed, May 11, 2005 at 11:00:34PM +0200, Achim Friedland wrote:
> Hello,
>=20
>=20
> I just tryed to compile a patched kernel-2.6.9 and/or kernel-2.6.11.8,=20
> but with both kernels I get alot of errors...
>=20
>=20
> CC [M]  net/ipv4/netfilter/ipt_spec_nh.o
>   CC [M]  net/ipv4/netfilter/ipt_MASQUERADE.o
>   CC [M]  net/ipv4/netfilter/ipt_REDIRECT.o
>   CC [M]  net/ipv4/netfilter/ipt_NETMAP.o
>   CC [M]  net/ipv4/netfilter/ipt_SAME.o
>   CC [M]  net/ipv4/netfilter/ipt_CLASSIFY.o
>   CC [M]  net/ipv4/netfilter/ip_nat_snmp_basic.o
>   CC [M]  net/ipv4/netfilter/ipt_LOG.o
>   CC [M]  net/ipv4/netfilter/ipt_ULOG.o
>   CC [M]  net/ipv4/netfilter/ipt_TCPMSS.o
>   CC [M]  net/ipv4/netfilter/ipt_NOTRACK.o
>   CC [M]  net/ipv4/netfilter/arp_tables.o
>   CC [M]  net/ipv4/netfilter/arpt_mangle.o
>   CC [M]  net/ipv4/netfilter/arptable_filter.o
>   CC [M]  net/ipv4/netfilter/ip_queue.o
>   CC [M]  net/ipv6/netfilter/ip6_tables.o
>   CC [M]  net/ipv6/netfilter/ip6t_limit.o
>   CC [M]  net/ipv6/netfilter/ip6t_mark.o
>   CC [M]  net/ipv6/netfilter/ip6t_length.o
>   CC [M]  net/ipv6/netfilter/ip6t_mac.o
>   CC [M]  net/ipv6/netfilter/ip6t_rt.o
>   CC [M]  net/ipv6/netfilter/ip6t_hbh.o
>   CC [M]  net/ipv6/netfilter/ip6t_dst.o
>   CC [M]  net/ipv6/netfilter/ip6t_ipv6header.o
>   CC [M]  net/ipv6/netfilter/ip6t_frag.o
>   CC [M]  net/ipv6/netfilter/ip6t_esp.o
>   CC [M]  net/ipv6/netfilter/ip6t_ah.o
>   CC [M]  net/ipv6/netfilter/ip6t_eui64.o
>   CC [M]  net/ipv6/netfilter/ip6t_multiport.o
>   CC [M]  net/ipv6/netfilter/ip6t_owner.o
>   CC [M]  net/ipv6/netfilter/ip6t_physdev.o
>   CC [M]  net/ipv6/netfilter/ip6table_filter.o
>   CC [M]  net/ipv6/netfilter/ip6table_mangle.o
>   CC [M]  net/ipv6/netfilter/ip6t_spec_nh.o
>   CC [M]  net/ipv6/netfilter/ip6t_MARK.o
>   CC [M]  net/ipv6/netfilter/ip6_queue.o
>   CC [M]  net/ipv6/netfilter/ip6t_LOG.o
>   CC [M]  net/ipv6/netfilter/ip6table_raw.o
>   CC [M]  net/ipv6/netfilter/ip6t_hl.o
>   CC [M]  net/mpls/af_mpls.o
>   CC [M]  net/mpls/mpls_if.o
>   CC [M]  net/mpls/mpls_ilm.o
>   CC [M]  net/mpls/mpls_init.o
>   CC [M]  net/mpls/mpls_input.o
>   CC [M]  net/mpls/mpls_opcode.o
>   CC [M]  net/mpls/mpls_nhlfe.o
>   CC [M]  net/mpls/mpls_output.o
>   CC [M]  net/mpls/mpls_utils.o
>   CC [M]  net/mpls/mpls_dst.o
>   CC [M]  net/mpls/mpls_netlink.o
>   CC [M]  net/mpls/mpls_proto.o
>   CC [M]  net/mpls/mpls_sysfs.o
>   CC [M]  net/mpls/mpls_instr.o
>   CC [M]  net/mpls/mpls_tunnel.o
>   LD [M]  net/mpls/mpls.o
> net/mpls/mpls_netlink.o(.init.text+0x0): In function `init_module':
> : multiple definition of `init_module'
> net/mpls/mpls_init.o(.init.text+0x0): first defined here
> ld: Warning: size of symbol `init_module' changed from 189 in=20
> net/mpls/mpls_init.o to 13 in net/mpls/mpls_netlink.o
> net/mpls/mpls_netlink.o(.exit.text+0x0): In function `cleanup_module':
> : multiple definition of `cleanup_module'
> net/mpls/mpls_init.o(.exit.text+0x0): first defined here
> ld: Warning: size of symbol `cleanup_module' changed from 97 in=20
> net/mpls/mpls_init.o to 11 in net/mpls/mpls_netlink.o
> net/mpls/mpls_tunnel.o(.exit.text+0x0): In function `cleanup_module':
> : multiple definition of `cleanup_module'
> net/mpls/mpls_init.o(.exit.text+0x0): first defined here
> ld: Warning: size of symbol `cleanup_module' changed from 11 in=20
> net/mpls/mpls_init.o to 70 in net/mpls/mpls_tunnel.o
> net/mpls/mpls_tunnel.o(.init.text+0x0): In function `init_module':
> : multiple definition of `init_module'
> net/mpls/mpls_init.o(.init.text+0x0): first defined here
> ld: Warning: size of symbol `init_module' changed from 13 in=20
> net/mpls/mpls_init.o to 76 in net/mpls/mpls_tunnel.o
> make[3]: *** [net/mpls/mpls.o] Error 1
> make[2]: *** [net/mpls] Error 2
> make[1]: *** [net] Error 2
> make[1]: Leaving directory=20
> `/home/ahzf/SOEKRIS/ap1-goLAN/usr/src/linux-2.6.9-soekris-elan-mpls'
> make: *** [stamp-build] Error 2
>=20
>=20
>=20
> Any suggestions how to solve this?
>=20
>=20
>=20
> achim
>=20
> PS: There are some minor problems while patching the 2.6.11.8 kernel,
>     but I think nothing really imporant...
>=20
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes
> Want to be the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_id=3D7393&alloc_id=3D16281&op=3Dclick
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--=20
James R. Leu
jl...@mi...

[mpls-linux-general] Problems with linux 2.6.9 or linux 2.6.11.8

[Achim F. <mai...@ah...>]

Hello,


I just tryed to compile a patched kernel-2.6.9 and/or kernel-2.6.11.8, 
but with both kernels I get alot of errors...


CC [M]  net/ipv4/netfilter/ipt_spec_nh.o
   CC [M]  net/ipv4/netfilter/ipt_MASQUERADE.o
   CC [M]  net/ipv4/netfilter/ipt_REDIRECT.o
   CC [M]  net/ipv4/netfilter/ipt_NETMAP.o
   CC [M]  net/ipv4/netfilter/ipt_SAME.o
   CC [M]  net/ipv4/netfilter/ipt_CLASSIFY.o
   CC [M]  net/ipv4/netfilter/ip_nat_snmp_basic.o
   CC [M]  net/ipv4/netfilter/ipt_LOG.o
   CC [M]  net/ipv4/netfilter/ipt_ULOG.o
   CC [M]  net/ipv4/netfilter/ipt_TCPMSS.o
   CC [M]  net/ipv4/netfilter/ipt_NOTRACK.o
   CC [M]  net/ipv4/netfilter/arp_tables.o
   CC [M]  net/ipv4/netfilter/arpt_mangle.o
   CC [M]  net/ipv4/netfilter/arptable_filter.o
   CC [M]  net/ipv4/netfilter/ip_queue.o
   CC [M]  net/ipv6/netfilter/ip6_tables.o
   CC [M]  net/ipv6/netfilter/ip6t_limit.o
   CC [M]  net/ipv6/netfilter/ip6t_mark.o
   CC [M]  net/ipv6/netfilter/ip6t_length.o
   CC [M]  net/ipv6/netfilter/ip6t_mac.o
   CC [M]  net/ipv6/netfilter/ip6t_rt.o
   CC [M]  net/ipv6/netfilter/ip6t_hbh.o
   CC [M]  net/ipv6/netfilter/ip6t_dst.o
   CC [M]  net/ipv6/netfilter/ip6t_ipv6header.o
   CC [M]  net/ipv6/netfilter/ip6t_frag.o
   CC [M]  net/ipv6/netfilter/ip6t_esp.o
   CC [M]  net/ipv6/netfilter/ip6t_ah.o
   CC [M]  net/ipv6/netfilter/ip6t_eui64.o
   CC [M]  net/ipv6/netfilter/ip6t_multiport.o
   CC [M]  net/ipv6/netfilter/ip6t_owner.o
   CC [M]  net/ipv6/netfilter/ip6t_physdev.o
   CC [M]  net/ipv6/netfilter/ip6table_filter.o
   CC [M]  net/ipv6/netfilter/ip6table_mangle.o
   CC [M]  net/ipv6/netfilter/ip6t_spec_nh.o
   CC [M]  net/ipv6/netfilter/ip6t_MARK.o
   CC [M]  net/ipv6/netfilter/ip6_queue.o
   CC [M]  net/ipv6/netfilter/ip6t_LOG.o
   CC [M]  net/ipv6/netfilter/ip6table_raw.o
   CC [M]  net/ipv6/netfilter/ip6t_hl.o
   CC [M]  net/mpls/af_mpls.o
   CC [M]  net/mpls/mpls_if.o
   CC [M]  net/mpls/mpls_ilm.o
   CC [M]  net/mpls/mpls_init.o
   CC [M]  net/mpls/mpls_input.o
   CC [M]  net/mpls/mpls_opcode.o
   CC [M]  net/mpls/mpls_nhlfe.o
   CC [M]  net/mpls/mpls_output.o
   CC [M]  net/mpls/mpls_utils.o
   CC [M]  net/mpls/mpls_dst.o
   CC [M]  net/mpls/mpls_netlink.o
   CC [M]  net/mpls/mpls_proto.o
   CC [M]  net/mpls/mpls_sysfs.o
   CC [M]  net/mpls/mpls_instr.o
   CC [M]  net/mpls/mpls_tunnel.o
   LD [M]  net/mpls/mpls.o
net/mpls/mpls_netlink.o(.init.text+0x0): In function `init_module':
: multiple definition of `init_module'
net/mpls/mpls_init.o(.init.text+0x0): first defined here
ld: Warning: size of symbol `init_module' changed from 189 in 
net/mpls/mpls_init.o to 13 in net/mpls/mpls_netlink.o
net/mpls/mpls_netlink.o(.exit.text+0x0): In function `cleanup_module':
: multiple definition of `cleanup_module'
net/mpls/mpls_init.o(.exit.text+0x0): first defined here
ld: Warning: size of symbol `cleanup_module' changed from 97 in 
net/mpls/mpls_init.o to 11 in net/mpls/mpls_netlink.o
net/mpls/mpls_tunnel.o(.exit.text+0x0): In function `cleanup_module':
: multiple definition of `cleanup_module'
net/mpls/mpls_init.o(.exit.text+0x0): first defined here
ld: Warning: size of symbol `cleanup_module' changed from 11 in 
net/mpls/mpls_init.o to 70 in net/mpls/mpls_tunnel.o
net/mpls/mpls_tunnel.o(.init.text+0x0): In function `init_module':
: multiple definition of `init_module'
net/mpls/mpls_init.o(.init.text+0x0): first defined here
ld: Warning: size of symbol `init_module' changed from 13 in 
net/mpls/mpls_init.o to 76 in net/mpls/mpls_tunnel.o
make[3]: *** [net/mpls/mpls.o] Error 1
make[2]: *** [net/mpls] Error 2
make[1]: *** [net] Error 2
make[1]: Leaving directory 
`/home/ahzf/SOEKRIS/ap1-goLAN/usr/src/linux-2.6.9-soekris-elan-mpls'
make: *** [stamp-build] Error 2



Any suggestions how to solve this?



achim

PS: There are some minor problems while patching the 2.6.11.8 kernel,
     but I think nothing really imporant...

Re: [mpls-linux-general] Mpls qos and segfault

[James R. L. <jl...@mi...>]

On Tue, May 10, 2005 at 11:02:28AM +0530, ram...@wi... wrote:
>=20
> I recompiled the kernel now it's working fine.
> Now I have one more question.
>=20
> mplsadm tool contains proc support.but in mpls tool in proc, only
> spech_nh is there.
> In sysfs/mpls directory I have only debug and version files. Where can I
> get label,nhlfe entries
> Other than mpls show command.

All of those 'files' have been removed.  You can now get the same info with:

    mpls ilm show
    mpls nhlfe show
    mpls labelspace show
    mpls xc show

>=20
> Thanks in advance,
>=20
> Regards
> Ramesh.s
>=20
> -----Original Message-----
> From: James R. Leu [mailto:jl...@mi...]
> Sent: Tuesday, May 10, 2005 12:25 AM
> To: Ramesh Selvaraj (WT01 - TELECOM SOLUTIONS)
> Cc: vin...@im...; mpl...@li...
> Subject: Re: [mpls-linux-general] Mpls qos and segfault
>=20
> Did you compile it in or as a module?  If it is as a module, you need to
> load the module.
>=20
> On Thu, May 05, 2005 at 09:15:00PM +0530, ram...@wi...
> wrote:
> > Hi
> >
> > I have chosen both MPLS support and MPLS IP supportand spec_nh also.
> >
> > Regards
> > Ramesh.s
> >
> > -----Original Message-----
> > From: Vincent Untz [mailto:vin...@im...]
> > Sent: Thursday, May 05, 2005 1:27 PM
> > To: Ramesh Selvaraj (WT01 - TELECOM SOLUTIONS)
> > Cc: mpl...@li...
> > Subject: Re: [mpls-linux-general] Mpls qos and segfault
> >
> > Hi,
> >
> > Le jeudi 05 mai 2005 ? 08:47 +0530, ram...@wi... a ?crit
> :
> > > i got segmentation fault while using mpls tool.
> > > i added labelspace as follows
> > >
> > > mpls labelspace add dev eth0 labelspace 0
> > >
> > > mpls labelspace show
> > >
> > > shows the labelspaces. after that the following ilm command gives
> > > segfault.
> > >
> > > mpls ilm add label gen 1000 labelspace 0
> > >
> > > dmesg says
> > >
> > >
> > > MPLS DEBUG net/mpls/mpls_ilm.c:146:mpls_ilm_dst_alloc: *Unable to
> > > find protocol driver for 'AF_INET'* MPLS DEBUG
> > > net/mpls/mpls_ilm.c:157:mpls_ilm_dst_alloc: exit Unable to handle
> > > kernel NULL pointer dereference at virtual address 00000000
> printing eip:
> > > c0322e34
> > > *pde =3D 00000000
> > > Oops: 0000 [#1]
> > > ..............
> > > ..................
> > >
> > > now what is the problem, how to rectify that error *Unable to find
> > > protocol driver for 'AF_INET'
> >
> > Did you compile the MPLS IP support in the kernel? Or just the MPLS
> support?
> >
> > IIRC, you need to activate MPLS support and then MPLS IP support when
> you configure the kernel (it's not done by default).
> >
> > Vincent
> >
> > --
> > Les gens heureux ne sont pas press?s.
> >
> >
> > -------------------------------------------------------
> > This SF.Net email is sponsored by: NEC IT Guy Games.
> > Get your fingers limbered up and give it your best shot. 4 great
> > events, 4 opportunities to win big! Highest score wins.NEC IT Guy
> > Games. Play to win an NEC 61 plasma display. Visit
> > http://www.necitguy.com/?r
> > _______________________________________________
> > mpls-linux-general mailing list
> > mpl...@li...
> > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
>=20
> --
> James R. Leu
> jl...@mi...
>=20
>=20
>=20
> Confidentiality Notice
>=20
> The information contained in this electronic message and any attachments =
to this message are intended
> for the exclusive use of the addressee(s) and may contain confidential or=
 privileged information. If
> you are not the intended recipient, please notify the sender at Wipro or =
Mai...@wi... immediately
> and destroy all copies of this message and any attachments.

--=20
James R. Leu
jl...@mi...

RE: [mpls-linux-general] Mpls qos and segfault

[Vincent U. <vin...@im...>]

Le mardi 10 mai 2005 =C3=A0 11:02 +0530, ram...@wi... a =C3=A9=
crit :
> I recompiled the kernel now it's working fine.
> Now I have one more question.
>=20
> mplsadm tool contains proc support.but in mpls tool in proc, only
> spech_nh is there.
> In sysfs/mpls directory I have only debug and version files. Where can I
> get label,nhlfe entries
> Other than mpls show command.

You can access this information via a netlink socket. The mpls command
source code is a good basis if you want to know how to do it.

But it might be easier to just parse the output of "mpls show"...

Vincent

--=20
Les gens heureux ne sont pas press=C3=A9s.

RE: [mpls-linux-general] Mpls qos and segfault

[<ram...@wi...>]

I recompiled the kernel now it's working fine.
Now I have one more question.

mplsadm tool contains proc support.but in mpls tool in proc, only
spech_nh is there.
In sysfs/mpls directory I have only debug and version files. Where can I
get label,nhlfe entries
Other than mpls show command.

Thanks in advance,

Regards
Ramesh.s

-----Original Message-----
From: James R. Leu [mailto:jl...@mi...]=20
Sent: Tuesday, May 10, 2005 12:25 AM
To: Ramesh Selvaraj (WT01 - TELECOM SOLUTIONS)
Cc: vin...@im...; mpl...@li...
Subject: Re: [mpls-linux-general] Mpls qos and segfault

Did you compile it in or as a module?  If it is as a module, you need to
load the module.

On Thu, May 05, 2005 at 09:15:00PM +0530, ram...@wi...
wrote:
> Hi
>=20
> I have chosen both MPLS support and MPLS IP supportand spec_nh also.
>=20
> Regards
> Ramesh.s
>=20
> -----Original Message-----
> From: Vincent Untz [mailto:vin...@im...]
> Sent: Thursday, May 05, 2005 1:27 PM
> To: Ramesh Selvaraj (WT01 - TELECOM SOLUTIONS)
> Cc: mpl...@li...
> Subject: Re: [mpls-linux-general] Mpls qos and segfault
>=20
> Hi,
>=20
> Le jeudi 05 mai 2005 ? 08:47 +0530, ram...@wi... a ?crit
:
> > i got segmentation fault while using mpls tool.
> > i added labelspace as follows
> >=20
> > mpls labelspace add dev eth0 labelspace 0
> >=20
> > mpls labelspace show
> >=20
> > shows the labelspaces. after that the following ilm command gives=20
> > segfault.
> >=20
> > mpls ilm add label gen 1000 labelspace 0
> >=20
> > dmesg says
> >=20
> >=20
> > MPLS DEBUG net/mpls/mpls_ilm.c:146:mpls_ilm_dst_alloc: *Unable to=20
> > find protocol driver for 'AF_INET'* MPLS DEBUG
> > net/mpls/mpls_ilm.c:157:mpls_ilm_dst_alloc: exit Unable to handle=20
> > kernel NULL pointer dereference at virtual address 00000000
printing eip:
> > c0322e34
> > *pde =3D 00000000
> > Oops: 0000 [#1]
> > ..............
> > ..................
> >=20
> > now what is the problem, how to rectify that error *Unable to find=20
> > protocol driver for 'AF_INET'
>=20
> Did you compile the MPLS IP support in the kernel? Or just the MPLS
support?
>=20
> IIRC, you need to activate MPLS support and then MPLS IP support when
you configure the kernel (it's not done by default).
>=20
> Vincent
>=20
> --
> Les gens heureux ne sont pas press?s.
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: NEC IT Guy Games.
> Get your fingers limbered up and give it your best shot. 4 great=20
> events, 4 opportunities to win big! Highest score wins.NEC IT Guy=20
> Games. Play to win an NEC 61 plasma display. Visit=20
> http://www.necitguy.com/?r=20
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--
James R. Leu
jl...@mi...

Re: [mpls-linux-general] Possible Iptables --spec_nh bug

[James R. L. <jl...@mi...>]

On Mon, May 02, 2005 at 09:34:34AM -0300, jrussi wrote:
> Leu,
>=20
> You see the output from iptables -L below? Shouldn't it be
> "set spec_nh 0x8847" instead of "set spec_nh 0x4788"?

Yes.  I must have missed a ntohs somewhere.  You will have to check the
kernel code and the userland code.  Let me know if you find it.
Otherwise I will put it on my todo list.

> Please confirm.=20
> Thank you.
> Josdeyvi
>=20
> [root@localhost ~]# mpls nhlfe add key 0
> [root@localhost ~]# mpls nhlfe change key 0x2 instructions
> push gen 10000 nexthop eth0 ipv4 10.0.1.2
> [root@localhost ~]# iptables -A OUTPUT -d 10.0.1.2 -j
> spec_nh --spec_nh 0x8847:0x2
>=20
> [root@localhost ~]# iptables -L Chain INPUT (policy ACCEPT)
> target     prot opt source               destination
> RH-Firewall-1-INPUT  all  --  anywhere             anywhere
>=20
> Chain FORWARD (policy ACCEPT)
> target     prot opt source               destination
> RH-Firewall-1-INPUT  all  --  anywhere             anywhere
>=20
> Chain OUTPUT (policy ACCEPT)
> target     prot opt source               destination
> spec_nh    all  --  anywhere             10.0.1.2         =20
>  set spec_nh 0x4788:0x00000002
>=20
>=20
>=20
> =20
> __________________________________________________________________________
> Acabe com aquelas janelinhas que pulam na sua tela.
> AntiPop-up UOL - ? gr?tis!
> http://antipopup.uol.com.br/
>=20
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: NEC IT Guy Games.
> Get your fingers limbered up and give it your best shot. 4 great events, 4
> opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
> win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--=20
James R. Leu
jl...@mi...

Re: [mpls-linux-general] Mpls qos and segfault

[James R. L. <jl...@mi...>]

Did you compile it in or as a module?  If it is as a module, you need to
load the module.

On Thu, May 05, 2005 at 09:15:00PM +0530, ram...@wi... wrote:
> Hi
>=20
> I have chosen both MPLS support and MPLS IP supportand spec_nh also.
>=20
> Regards
> Ramesh.s
>=20
> -----Original Message-----
> From: Vincent Untz [mailto:vin...@im...]=20
> Sent: Thursday, May 05, 2005 1:27 PM
> To: Ramesh Selvaraj (WT01 - TELECOM SOLUTIONS)
> Cc: mpl...@li...
> Subject: Re: [mpls-linux-general] Mpls qos and segfault
>=20
> Hi,
>=20
> Le jeudi 05 mai 2005 ? 08:47 +0530, ram...@wi... a ?crit :
> > i got segmentation fault while using mpls tool.
> > i added labelspace as follows
> >=20
> > mpls labelspace add dev eth0 labelspace 0
> >=20
> > mpls labelspace show
> >=20
> > shows the labelspaces. after that the following ilm command gives=20
> > segfault.
> >=20
> > mpls ilm add label gen 1000 labelspace 0
> >=20
> > dmesg says
> >=20
> >=20
> > MPLS DEBUG net/mpls/mpls_ilm.c:146:mpls_ilm_dst_alloc: *Unable to find=
=20
> > protocol driver for 'AF_INET'* MPLS DEBUG
> > net/mpls/mpls_ilm.c:157:mpls_ilm_dst_alloc: exit Unable to handle=20
> > kernel NULL pointer dereference at virtual address 00000000  printing e=
ip:
> > c0322e34
> > *pde =3D 00000000
> > Oops: 0000 [#1]
> > ..............
> > ..................
> >=20
> > now what is the problem, how to rectify that error *Unable to find=20
> > protocol driver for 'AF_INET'
>=20
> Did you compile the MPLS IP support in the kernel? Or just the MPLS suppo=
rt?
>=20
> IIRC, you need to activate MPLS support and then MPLS IP support when you=
 configure the kernel (it's not done by default).
>=20
> Vincent
>=20
> --
> Les gens heureux ne sont pas press?s.
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: NEC IT Guy Games.
> Get your fingers limbered up and give it your best shot. 4 great events, 4
> opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
> win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general

--=20
James R. Leu
jl...@mi...