mpls-linux-general Mailing List for MPLS for Linux (Page 104)
Status: Beta
Brought to you by:
jleu
Menu
▾
▴
mpls-linux-general — Mailing list for general discussion about MPLS for Linux
You can subscribe to this list here.
| 2000 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(3) |
Dec
(26) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2001 |
Jan
(22) |
Feb
(19) |
Mar
(19) |
Apr
(45) |
May
(52) |
Jun
(101) |
Jul
(79) |
Aug
(24) |
Sep
(43) |
Oct
(54) |
Nov
(71) |
Dec
(53) |
| 2002 |
Jan
(111) |
Feb
(123) |
Mar
(67) |
Apr
(61) |
May
(75) |
Jun
(26) |
Jul
(36) |
Aug
(41) |
Sep
(79) |
Oct
(85) |
Nov
(58) |
Dec
(39) |
| 2003 |
Jan
(26) |
Feb
(61) |
Mar
(80) |
Apr
(56) |
May
(39) |
Jun
(44) |
Jul
(28) |
Aug
(25) |
Sep
(4) |
Oct
(20) |
Nov
(38) |
Dec
(9) |
| 2004 |
Jan
(14) |
Feb
(14) |
Mar
(68) |
Apr
(17) |
May
(45) |
Jun
(42) |
Jul
(41) |
Aug
(23) |
Sep
(46) |
Oct
(89) |
Nov
(55) |
Dec
(33) |
| 2005 |
Jan
(74) |
Feb
(39) |
Mar
(105) |
Apr
(96) |
May
(43) |
Jun
(48) |
Jul
(21) |
Aug
(22) |
Sep
(33) |
Oct
(28) |
Nov
(29) |
Dec
(81) |
| 2006 |
Jan
(37) |
Feb
(32) |
Mar
(147) |
Apr
(37) |
May
(33) |
Jun
(28) |
Jul
(15) |
Aug
(20) |
Sep
(15) |
Oct
(23) |
Nov
(30) |
Dec
(40) |
| 2007 |
Jan
(20) |
Feb
(24) |
Mar
(65) |
Apr
(69) |
May
(41) |
Jun
(53) |
Jul
(39) |
Aug
(76) |
Sep
(53) |
Oct
(43) |
Nov
(26) |
Dec
(24) |
| 2008 |
Jan
(19) |
Feb
(67) |
Mar
(91) |
Apr
(75) |
May
(47) |
Jun
(63) |
Jul
(68) |
Aug
(39) |
Sep
(44) |
Oct
(33) |
Nov
(62) |
Dec
(84) |
| 2009 |
Jan
(14) |
Feb
(39) |
Mar
(55) |
Apr
(63) |
May
(16) |
Jun
(9) |
Jul
(4) |
Aug
(6) |
Sep
(1) |
Oct
(2) |
Nov
(10) |
Dec
(5) |
| 2010 |
Jan
(3) |
Feb
(1) |
Mar
(5) |
Apr
(13) |
May
(4) |
Jun
(5) |
Jul
(2) |
Aug
(8) |
Sep
(6) |
Oct
(1) |
Nov
(2) |
Dec
(2) |
| 2011 |
Jan
(1) |
Feb
(21) |
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
(3) |
Aug
(6) |
Sep
|
Oct
|
Nov
(2) |
Dec
(6) |
| 2012 |
Jan
(5) |
Feb
(3) |
Mar
|
Apr
|
May
|
Jun
(1) |
Jul
(5) |
Aug
(3) |
Sep
(6) |
Oct
|
Nov
|
Dec
|
| 2013 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
|
| 2014 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2015 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(2) |
Dec
|
|
From: itrat q. <iqu...@ho...> - 2004-03-07 17:43:34
|
<html><div style='background-color:'><DIV class=RTE>After I carry out the following command<BR><BR>ip route add 11.0.1.2/32 via 11.0.1.2 lsp 0x2<BR><BR>and then do "cat /proc/net/mpls_out"<BR><BR>I get "0x00000002 0/0/0 1 PUSH(gen 10000) SET(eth1,255.255.255.255)" <BR><BR>instead of "0x00000002 0/0/0 1 PUSH(gen 10000) SET(eth1,11.0.1.2)"<BR><BR>Should I change the mpls_out file myself or is there some way to fix it?<BR><BR>Waiting for ur help.</DIV></div><br clear=all><hr>STOP MORE SPAM with <a href="http://g.msn.com/8HMBEN/2728??PS=">the new MSN 8</a> and get 2 months FREE*</html> |
|
From: James R. L. <jl...@mi...> - 2004-03-05 20:58:02
|
I wouldn't recommend getting the _whole_ depot. It is over 1G in size and
my ISP limits my upstream to 128Kbps :-)
I suggest making your view look like:
//depot/iproute2-mpls/... //<client name>/iproute2-mpls/...
//depot/iptables-mpls/... //<client name>/iptables-mpls/...
//depot/mpls-kernel/... //<client name>/mpls-kernel/...
//depot/mpls-linux/... //<client name>/mpls-linux/...
//depot/ldp-portable/... //<client name>/ldp-portable/...
//depot/quagga-ldp/... //<client name>/quagga-ldp/...
On Fri, Mar 05, 2004 at 08:34:35PM +0100, ar...@ch... wrote:
> It seems, that i have a most of your repository:
> #:/usr/src/MPLS/p4client# ls
> fireplace iptables-vrf linux-kernel-2.4
> mpls-linux-1.1
> iproute2 iputils linux-vrf ppp
> iproute2-davem iputils-vrf misc-tools ppp-mpls
> iproute2-mpls l2cc-kernel mpls-kernel quagga
> iproute2-mpls-1.1 l2cc-kernel-on-2.4 mpls-kernel-1.1 quagga-ldp
> iproute2-vrf l2cc-linux mpls-kernel-1.1-on-2.4 quagga-mpls
> iptables ldp-portable mpls-kernel-davem quagga-plus
> iptables-mpls linux-kernel mpls-linux uml-kernel
>
>
> Now, i see that in quagga-ldp there are some differences than in oryginal
> dist. {
> < show_lsa_summary (struct vty *vty, struct ospf_lsa *lsa, int self, int
> *sum)
> ---
> > show_lsa_summary (struct vty *vty, struct ospf_lsa *lsa, int self)
> } and others - it was ospf_vty.c
quagga-ldp is a branch from quagga-plus. quagga-plus is a branch from
quagga. quagga-mpls is also a branch from quagga-plus.
quagga-plus is where I make my quagga bug fixes that I submit back to the
quagga team. quagga-mpls is where I'm doing the majority of my work these
days.
> i cant see there any patching scripts for quagga tree.
>
> So should i replace them all (oryginal dist, or your dist or which -
> quagga, quagga-plus?) from quagga-ldp files ?
>
> is there some intruction about what is what in this repository ?
No.
> Is there need for RSVP package ?
No.
> Ok, that will we talk later, now i just want to start it ;)
>
>
>
> Arkadiusz Binder | routing is anywhere
> -----BEGIN GEEK CODE BLOCK-----
> Version: 3.12
> GCC/E/IT/S d s+: a-- C+++ UL+++>+$ P+ L+++ E--- W+++ N+ o-- K- w--
> O+ M- V- PS++ PE++ Y PGP- t- 5? X+ R- tv- b++ DI? D+++
> G+ e h++(----) r+++ y+++
> ------END GEEK CODE BLOCK------
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
--
James R. Leu
jl...@mi...
|
|
From: <ar...@ch...> - 2004-03-05 19:55:30
|
It seems, that i have a most of your repository:
#:/usr/src/MPLS/p4client# ls
fireplace iptables-vrf linux-kernel-2.4
mpls-linux-1.1
iproute2 iputils linux-vrf ppp
iproute2-davem iputils-vrf misc-tools ppp-mpls
iproute2-mpls l2cc-kernel mpls-kernel quagga
iproute2-mpls-1.1 l2cc-kernel-on-2.4 mpls-kernel-1.1 quagga-ldp
iproute2-vrf l2cc-linux mpls-kernel-1.1-on-2.4 quagga-mpls
iptables ldp-portable mpls-kernel-davem quagga-plus
iptables-mpls linux-kernel mpls-linux uml-kernel
Now, i see that in quagga-ldp there are some differences than in oryginal
dist. {
< show_lsa_summary (struct vty *vty, struct ospf_lsa *lsa, int self, int
*sum)
---
> show_lsa_summary (struct vty *vty, struct ospf_lsa *lsa, int self)
} and others - it was ospf_vty.c
i cant see there any patching scripts for quagga tree.
So should i replace them all (oryginal dist, or your dist or which -
quagga, quagga-plus?) from quagga-ldp files ?
is there some intruction about what is what in this repository ?
Is there need for RSVP package ?
Ok, that will we talk later, now i just want to start it ;)
Arkadiusz Binder | routing is anywhere
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCC/E/IT/S d s+: a-- C+++ UL+++>+$ P+ L+++ E--- W+++ N+ o-- K- w--
O+ M- V- PS++ PE++ Y PGP- t- 5? X+ R- tv- b++ DI? D+++
G+ e h++(----) r+++ y+++
------END GEEK CODE BLOCK------
|
|
From: James R. L. <jl...@mi...> - 2004-03-02 15:50:13
|
mpls-linux has 'instructions' which can set tcindex and nfmark on a packet. You should be able to use them to influence the QoS features of the linux networking stack. I would suggest understanding how linux can assert QoS on standard IPv4 packets first, then use that knowledge to make it work with MPLS. On Tue, Mar 02, 2004 at 06:57:29PM +1100, Jon...@cs... wrote: > Dear all, >=20 > Yet another newbie question... >=20 > What would be the best way to shape an LSP? >=20 > I did some search in the web and got some hits pointing other Linux mpls > implementations... Some people suggested to use tc filter u32 to do the > job while other did some patch to tc. Is there any success story in this > mpls-linux implementation?? >=20 > Thanks for your information... >=20 > Cheers, > Jonathan >=20 >=20 > ------------------------------------------------------- > SF.Net is sponsored by: Speed Start Your Linux Apps Now. > Build and deploy apps & Web services for Linux with > a free DVD software kit from IBM. Click Now! > http://ads.osdn.com/?ad_id=1356&alloc_id438&op?k > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu jl...@mi... |
|
From: James R. L. <jl...@mi...> - 2004-03-02 15:34:09
|
Ther most recent release (based on a 2.6 kernel) has significant support for IPv6, but it has not been tested. If you are willing to test it, I'm will to do bug fixes :-) On Tue, Mar 02, 2004 at 03:03:37PM +0000, Ricardo Pedro wrote: > HI! > > I want to know if this release of MPLS is capable of using the IPv6 Stack. > > I'm doing a master thesis with MPLS and IPv6 and would like to know if > it is currently usable MPLS-Linux with IPv6. > > Thank you in advance > Ricardo Pedro > > > ------------------------------------------------------- > SF.Net is sponsored by: Speed Start Your Linux Apps Now. > Build and deploy apps & Web services for Linux with > a free DVD software kit from IBM. Click Now! > http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu jl...@mi... |
|
From: Ricardo P. <ri...@so...> - 2004-03-02 15:20:34
|
HI! I want to know if this release of MPLS is capable of using the IPv6 Stack. I'm doing a master thesis with MPLS and IPv6 and would like to know if it is currently usable MPLS-Linux with IPv6. Thank you in advance Ricardo Pedro |
|
From: <Jon...@cs...> - 2004-03-02 08:10:16
|
Dear all, Yet another newbie question... What would be the best way to shape an LSP? I did some search in the web and got some hits pointing other Linux mpls implementations... Some people suggested to use tc filter u32 to do the job while other did some patch to tc. Is there any success story in this mpls-linux implementation?? Thanks for your information... Cheers, Jonathan |
|
From: James R. L. <jl...@mi...> - 2004-02-19 03:57:37
|
There currently is not an instruction to do this. It is not hard to add (if you coded it up I would include it). All you have to do is set skb->vrf. There is a set_rx intructions that has not been exposed to userland. You could try modifing mplsadm.c to add support for set_rx. Jim On Wed, Feb 18, 2004 at 04:42:59PM +1100, Jon...@cs... wrote: > Dear all, >=20 > It would be much appreciated if you could help me out with these two > (newbie??) questions... >=20 > (1) Using mpls label to map incoming traffic into different vrf: >=20 > I haven't got linux-vrf-0.100 installed but was trying to use "standard" > policy routing in linux (e.g. ip rule add from ip_addr_A dev if_B table > vrf_C) to put incoming traffic into different VRFs. >=20 > Is it possible to use an mpls label (e.g. if_B =3D mpls0) to do the job?? > In the email archive 19/07/2001 subject "MPLS-Linux", James stated > "remember MPLS tunnel interfaces are unidirectional ... They only do TX > no RX". Does it means that I can't do something like "mplsadm -A -B -I > gen:123:0 -T mpls0" to make a mpls tunnel for incoming traffic?? >=20 > Otherwise, is there a way to map mpls label into different routing table > instances (e.g. using linux-vrf)?? >=20 > (2) Interface belonging to multiple labelspaces: >=20 > Is it possible to assign multiple labelspaces to an interface?? Or what > I am trying to do is again the original purpose of labelspace ;o) >=20 > Thanks for your considerations and valuable time... >=20 > Cheers, > Jonathan >=20 > =20 >=20 >=20 > ------------------------------------------------------- > SF.Net is sponsored by: Speed Start Your Linux Apps Now. > Build and deploy apps & Web services for Linux with > a free DVD software kit from IBM. Click Now! > http://ads.osdn.com/?ad_id=1356&alloc_id438&op?k > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu jl...@mi... |
|
From: <Jon...@cs...> - 2004-02-18 05:47:49
|
Dear all, It would be much appreciated if you could help me out with these two (newbie??) questions... (1) Using mpls label to map incoming traffic into different vrf: I haven't got linux-vrf-0.100 installed but was trying to use "standard" policy routing in linux (e.g. ip rule add from ip_addr_A dev if_B table vrf_C) to put incoming traffic into different VRFs. Is it possible to use an mpls label (e.g. if_B =3D mpls0) to do the = job?? In the email archive 19/07/2001 subject "MPLS-Linux", James stated "remember MPLS tunnel interfaces are unidirectional ... They only do TX no RX". Does it means that I can't do something like "mplsadm -A -B -I gen:123:0 -T mpls0" to make a mpls tunnel for incoming traffic?? Otherwise, is there a way to map mpls label into different routing table instances (e.g. using linux-vrf)?? (2) Interface belonging to multiple labelspaces: Is it possible to assign multiple labelspaces to an interface?? Or what I am trying to do is again the original purpose of labelspace ;o) Thanks for your considerations and valuable time... Cheers, Jonathan =20 |
|
From: Rudy S W. <rs...@an...> - 2004-02-09 20:47:31
|
Hello,
I've downloaded the new ldp-portable and zebra-ldp from James' p4 server,
then recompiled with the zebra/ldp with both new versions (installed
mpls-linux-1.1, iptables, iproute2, ppp, glibc, with rpm from mpls
homepage). I'm trying to test the ldp label distribution of ldp-portable.
My topology is just two 2.4.18 red hat machines directly connected. I can
run zebra and ospfd daemons and see ospf neigbors.
Currently:
on comp A:
uml-1-ospf# sho ip ospf nei
Neighbor ID Pri State Dead Time Address Interface
RXmtL RqstL DBsmL
192.168.0.2 1 Full/DR 00:00:34 10.3.3.6 eth1:10.3.3.5
0 0 0
on comp B:
uml-1-ospf# sho ip ospf nei
Neighbor ID Pri State Dead Time Address Interface
RXmtL RqstL DBsmL
192.168.0.2 1 Full/Backup 00:00:33 10.3.3.5 eth1:10.3.3.6
0 0 0
My goal is to start an ldp session and see the LDP peers exchange labels.
After I start mplsd I can see both comp A and comp B see ldp neighbors
with "show ldp neighbor".
However, on comp A machine the debug messages come up with:
on Comp A
connect: Network is unreachable
OUT: ldp_session_attempt_setp: MPLS_FAILURE
On comp A
uml-1#show ldp session
2 192.168.0.2 0 NON-EXIST
but for comp B
there's no MPLS_FAILURE
uml-2#show ldp session
no established sessions
When I run mplsd with "mpls ldp" in the mplsd.conf, it gives me a
segmentation fault, and i ran it in gdb looks like vty is looking for an
interface.
I start mplsd in gdb and run it without the "mpls ldp in the mplsd.conf,
then i manually enable mpls ldp by telnet to mplsd. After that I can see
that both machines can see each other with "show ldp neighbor". However,
when i try to show ldp discovery, sometimes it seg faults with this
message:
in ldp_cfg_entity_get.... at ldp_cfg.c:341
341 e->adj_index = a ? a->index : 0;
I'm sorry there's so much in this email, I've been struggling to get mpls
up and running. Thanks for any help you can give me.
~rsw
|
|
From: Rudy S W. <rs...@an...> - 2004-02-06 19:25:28
|
Hello, I've been trying to re-compile zebra/ldp-portable with the new source from p4. However, i'm comparing the old ldp-portable source and the new one from p4 and don't see any instance of ldp_fib_callback. This is the compilation error i see on the new version when i apply the ldp patch to zebra-0.94 (mpls.h:38 parse error before "ldp_fib_callback") This type was defined in ldp_struct.h in the old version but seems to be missing from ldp_struct.h in the one i downloaded from p4. Is this a problem or am I doing something wrong? Is anyone else having this problem? Thanks for any help you can give me. ~rsw |
|
From: priya S. <mpl...@ya...> - 2004-02-05 20:51:17
|
hello I am novice to MPLS....I was trying to install the package ....i patched the necsarry file for iproute2.....when i tried to compile the iproute2 ...i encountered the following error....i am using Kernel 2.4.1.. Plz help me solve the problem.....I had written abt ths bfor...but no response....i hope for a response this time... Regards Priya My error [root@localhost src]# pwd /root/usr/src [root@localhost src]# cd iproute2 [root@localhost iproute2]# make make[1]: Entering directory `/root/usr/src/iproute2/lib' gcc -D_GNU_SOURCE -O2 -Wstrict-prototypes -Wall -g -I../include-glibc -include ../include-glibc/glibc-bugs.h -I/usr/src/linux/include -I../include -DRESOLVE_HOSTNAMES -c -o ll_proto.o ll_proto.c ll_proto.c:36: `ETH_P_ECHO' undeclared here (not in a function) ll_proto.c:36: initializer element is not constant ll_proto.c:36: (near initialization for `llproto_names[1].id') ll_proto.c:36: initializer element is not constant ll_proto.c:36: (near initialization for `llproto_names[1]') ll_proto.c:37: initializer element is not constant ll_proto.c:37: (near initialization for `llproto_names[2]') ll_proto.c:38: initializer element is not constant ll_proto.c:38: (near initialization for `llproto_names[3]') ll_proto.c:39: initializer element is not constant ll_proto.c:39: (near initialization for `llproto_names[4]') ll_proto.c:40: initializer element is not constant ll_proto.c:40: (near initialization for `llproto_names[5]') ll_proto.c:41: initializer element is not constant ll_proto.c:41: (near initialization for `llproto_names[6]') ll_proto.c:42: initializer element is not constant ll_proto.c:42: (near initialization for `llproto_names[7]') ll_proto.c:43: initializer element is not constant ll_proto.c:43: (near initialization for `llproto_names[8]') ll_proto.c:44: initializer element is not constant ll_proto.c:44: (near initialization for `llproto_names[9]') ll_proto.c:45: initializer element is not constant ll_proto.c:45: (near initialization for `llproto_names[10]') ll_proto.c:46: initializer element is not constant ll_proto.c:46: (near initialization for `llproto_names[11]') ll_proto.c:47: initializer element is not constant ll_proto.c:47: (near initialization for `llproto_names[12]') ll_proto.c:48: initializer element is not constant ll_proto.c:48: (near initialization for `llproto_names[13]') ll_proto.c:49: initializer element is not constant ll_proto.c:49: (near initialization for `llproto_names[14]') ll_proto.c:50: initializer element is not constant ll_proto.c:50: (near initialization for `llproto_names[15]') ll_proto.c:51: initializer element is not constant ll_proto.c:51: (near initialization for `llproto_names[16]') ll_proto.c:52: initializer element is not constant ll_proto.c:52: (near initialization for `llproto_names[17]') ll_proto.c:53: initializer element is not constant ll_proto.c:53: (near initialization for `llproto_names[18]') ll_proto.c:54: initializer element is not constant ll_proto.c:54: (near initialization for `llproto_names[19]') ll_proto.c:55: initializer element is not constant ll_proto.c:55: (near initialization for `llproto_names[20]') ll_proto.c:56: initializer element is not constant ll_proto.c:56: (near initialization for `llproto_names[21]') ll_proto.c:57: initializer element is not constant ll_proto.c:57: (near initialization for `llproto_names[22]') ll_proto.c:58: initializer element is not constant ll_proto.c:58: (near initialization for `llproto_names[23]') ll_proto.c:59: initializer element is not constant ll_proto.c:59: (near initialization for `llproto_names[24]') ll_proto.c:60: initializer element is not constant ll_proto.c:60: (near initialization for `llproto_names[25]') ll_proto.c:61: initializer element is not constant ll_proto.c:61: (near initialization for `llproto_names[26]') ll_proto.c:62: initializer element is not constant ll_proto.c:62: (near initialization for `llproto_names[27]') ll_proto.c:63: initializer element is not constant ll_proto.c:63: (near initialization for `llproto_names[28]') ll_proto.c:64: initializer element is not constant ll_proto.c:64: (near initialization for `llproto_names[29]') ll_proto.c:65: initializer element is not constant ll_proto.c:65: (near initialization for `llproto_names[30]') ll_proto.c:66: initializer element is not constant ll_proto.c:66: (near initialization for `llproto_names[31]') ll_proto.c:67: initializer element is not constant ll_proto.c:67: (near initialization for `llproto_names[32]') ll_proto.c:68: initializer element is not constant ll_proto.c:68: (near initialization for `llproto_names[33]') ll_proto.c:70: initializer element is not constant ll_proto.c:70: (near initialization for `llproto_names[34]') ll_proto.c:71: initializer element is not constant ll_proto.c:71: (near initialization for `llproto_names[35]') make[1]: *** [ll_proto.o] Error 1 make[1]: Leaving directory `/root/usr/src/iproute2/lib' make: *** [all] Error 2 . --------------------------------- Do you Yahoo!? Yahoo! Finance: Get your refund fast by filing online |
|
From: James R. L. <jl...@mi...> - 2004-02-03 02:56:14
|
On Mon, Feb 02, 2004 at 06:44:19PM +0100, Arturo P. Gonzalez Herranz wrote: > James R. Leu escribi?: > >On Mon, Feb 02, 2004 at 02:21:39PM +0100, Arturo P. Gonzalez Herranz wrote: > > > >>Hi all, > >> I've been working for some months with MPLS, I've tried several > >>configurations, working only with MPLS boxes and interconnenting Cisco > >>routers with Linux boxes (with ripd and mplsd), it works really good. > >>Now I'm working with BGP/MPLS VPNs (RFC 2547) with Cisco routers and I'd > >>like to know what about of supporting BGP/MPLS VPNs is in this project. > >>I mean, if it would be very hard to add this funcionality to zebra and > >>if someone it's working on it. > > > > > >I have done some initial investigation as to how this can be implemented. > >It is possible but a good amount of infrastructure needs to be put in > >place. > >The combination of my linux-vrf project (http://linux-vrf.sf.net/) and > >my mpls-linux project provide the kernel infrastructure needed. The next > >step is to make some infrastructure changes to zebra/quagga. The first > >major step is to move the MPLS LSR code into the main zebra daemon which > >will require a new set of zebra protocol messages to be implemented. > >Next the CLI needs to be implemented and tied to the VRF and MPLS code. > >After that is all protocol interation which needs to be figured out. > > > >I have had people show interest in helping out on this project but I have > >not > >seen any code produced. At some point I will switch my concentration back > >to > >the protocol side of things (Ramon and myself have been concentrating on > >re-vamping the MPLS forwarding plane implementation). > > > >If your interesting is helping out, I can point you in the right direction. > > > > > >> Thanks a lot, Art > >> > >> > >> > >> > >>------------------------------------------------------- > >>The SF.Net email is sponsored by EclipseCon 2004 > >>Premiere Conference on Open Tools Development and Integration > >>See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. > >>http://www.eclipsecon.org/osdn > >>_______________________________________________ > >>mpls-linux-general mailing list > >>mpl...@li... > >>https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > > > > > Hi all again, > I see, there is still a lot work to do, isn't it? Anyway if had > some time I'd like to do some work in the right direction. So James, > could you explain in some detail which would the nexts steps to follow. The first step is to implement the MPLS LSR-MIB functionality in zebra. I already have a 'portable' LSR implementation (part of ldp-portable) which could be used as the base. It already has a zebra CLI implementation too. So it would pretty much be cutting the code from mplsd and adding it to zebra (quagga). Then new zebra messages for add/del in/out segments and add/del cross-connects need to be designed and implemented. Once that is in place then BGP and LDP need to be modified to use the new messages to install labeling info. At that point then the details of how MPLS label info interacts with the routing table and bgp peers would need to be worked out. It seems like a lot of work ... well because it is. The first step of moving the LSR code to zebra is not too difficult, and would be a good way to learn about the zebra code. > > Thanks, Art > > > > > ------------------------------------------------------- > The SF.Net email is sponsored by EclipseCon 2004 > Premiere Conference on Open Tools Development and Integration > See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. > http://www.eclipsecon.org/osdn > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu jl...@mi... |
|
From: Arturo P. G. H. <apg...@it...> - 2004-02-02 17:44:54
|
James R. Leu escribi=F3: > On Mon, Feb 02, 2004 at 02:21:39PM +0100, Arturo P. Gonzalez Herranz wr= ote: >=20 >>Hi all, >> I've been working for some months with MPLS, I've tried several=20 >>configurations, working only with MPLS boxes and interconnenting Cisco=20 >>routers with Linux boxes (with ripd and mplsd), it works really good.=20 >>Now I'm working with BGP/MPLS VPNs (RFC 2547) with Cisco routers and I'= d=20 >>like to know what about of supporting BGP/MPLS VPNs is in this project.= =20 >>I mean, if it would be very hard to add this funcionality to zebra and=20 >>if someone it's working on it. >=20 >=20 > I have done some initial investigation as to how this can be implemente= d. > It is possible but a good amount of infrastructure needs to be put in p= lace. > The combination of my linux-vrf project (http://linux-vrf.sf.net/) and > my mpls-linux project provide the kernel infrastructure needed. The ne= xt > step is to make some infrastructure changes to zebra/quagga. The first > major step is to move the MPLS LSR code into the main zebra daemon whic= h > will require a new set of zebra protocol messages to be implemented. > Next the CLI needs to be implemented and tied to the VRF and MPLS code. > After that is all protocol interation which needs to be figured out. >=20 > I have had people show interest in helping out on this project but I ha= ve not > seen any code produced. At some point I will switch my concentration b= ack to > the protocol side of things (Ramon and myself have been concentrating o= n > re-vamping the MPLS forwarding plane implementation). >=20 > If your interesting is helping out, I can point you in the right direct= ion. >=20 >=20 >> Thanks a lot, Art >> >> >> >> >>------------------------------------------------------- >>The SF.Net email is sponsored by EclipseCon 2004 >>Premiere Conference on Open Tools Development and Integration >>See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. >>http://www.eclipsecon.org/osdn >>_______________________________________________ >>mpls-linux-general mailing list >>mpl...@li... >>https://lists.sourceforge.net/lists/listinfo/mpls-linux-general >=20 >=20 Hi all again, I see, there is still a lot work to do, isn't it? Anyway if had=20 some time I'd like to do some work in the right direction. So James,=20 could you explain in some detail which would the nexts steps to follow. Thanks, Art |
|
From: James R. L. <jl...@mi...> - 2004-02-02 15:04:18
|
On Mon, Feb 02, 2004 at 02:21:39PM +0100, Arturo P. Gonzalez Herranz wrote: > Hi all, > I've been working for some months with MPLS, I've tried several > configurations, working only with MPLS boxes and interconnenting Cisco > routers with Linux boxes (with ripd and mplsd), it works really good. > Now I'm working with BGP/MPLS VPNs (RFC 2547) with Cisco routers and I'd > like to know what about of supporting BGP/MPLS VPNs is in this project. > I mean, if it would be very hard to add this funcionality to zebra and > if someone it's working on it. I have done some initial investigation as to how this can be implemented. It is possible but a good amount of infrastructure needs to be put in place. The combination of my linux-vrf project (http://linux-vrf.sf.net/) and my mpls-linux project provide the kernel infrastructure needed. The next step is to make some infrastructure changes to zebra/quagga. The first major step is to move the MPLS LSR code into the main zebra daemon which will require a new set of zebra protocol messages to be implemented. Next the CLI needs to be implemented and tied to the VRF and MPLS code. After that is all protocol interation which needs to be figured out. I have had people show interest in helping out on this project but I have not seen any code produced. At some point I will switch my concentration back to the protocol side of things (Ramon and myself have been concentrating on re-vamping the MPLS forwarding plane implementation). If your interesting is helping out, I can point you in the right direction. > Thanks a lot, Art > > > > > ------------------------------------------------------- > The SF.Net email is sponsored by EclipseCon 2004 > Premiere Conference on Open Tools Development and Integration > See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. > http://www.eclipsecon.org/osdn > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu jl...@mi... |
|
From: Ramon C. <cas...@in...> - 2004-02-02 13:46:12
|
On Mon, 2 Feb 2004, Arturo P. Gonzalez Herranz wrote: > Hi all, > I've been working for some months with MPLS, I've tried several > configurations, working only with MPLS boxes and interconnenting Cisco > routers with Linux boxes (with ripd and mplsd), it works really good. > Now I'm working with BGP/MPLS VPNs (RFC 2547) with Cisco routers and I'd > like to know what about of supporting BGP/MPLS VPNs is in this project. > I mean, if it would be very hard to add this funcionality to zebra and > if someone it's working on it. Some work is being done (mostly by James). You should check his p4 repository, and also the quagaa project web file. I am also interested in setting experimental MPLS/BGP VPNs + L2 XC, an dI plan to jump in as soon as we release the new kernel patch for 2.6. Linux has kernel support for several tables, but maybe James could elaborate an overview / list of pending tasks to start with. regards, Ramon |
|
From: Arturo P. G. H. <apg...@it...> - 2004-02-02 13:22:13
|
Hi all,
I've been working for some months with MPLS, I've tried several
configurations, working only with MPLS boxes and interconnenting Cisco
routers with Linux boxes (with ripd and mplsd), it works really good.
Now I'm working with BGP/MPLS VPNs (RFC 2547) with Cisco routers and I'd
like to know what about of supporting BGP/MPLS VPNs is in this project.
I mean, if it would be very hard to add this funcionality to zebra and
if someone it's working on it.
Thanks a lot, Art
|
|
From: James R. L. <jl...@mi...> - 2004-02-01 21:19:18
|
Get the latest ldp-portable from p4 (look at http://mpls-linux.sf.net/ for howto info). The latest version fixes many bugs (including the one you are experiencing) On Sun, Feb 01, 2004 at 03:03:33PM -0500, Rudy S Wen wrote: > Hello, > > I've setup a point to point connection between two linux red hat 2.4.18 > machines (installed mpls/ldp through the rpms). I'm trying to setup up an > LDP session between these two machines. Zebra and ospfd are both > running and I can see an ospf adjacency (telnet to ospfd, to run show ip > ospf neighbor). Then I invoke mplsd on both machines with a mpld.conf > file where i setup mpls ldp on eth1 (for both machines). When i log into > mplsd and try the show ldp neighbor command, mplsd gets a seg fault.( I > can run > show ldp session, database, discovery with no problem). This only seems > to happen when the other mpls daemon is running. Am I doing something > wrong? Here are my .conf files > > > -----zebra.conf------ > hostname host1-zebra > password root > enable password root > service advanced-vty > ! > interface lo > ! > interface dummy0 > ip address 192.168.0.1/32 > ! > interface eth0 > ip address 10.3.3.1/30 > ! > interface eth1 > ip address 10.3.3.5/30 > ! > line vty > exec-timeout 0 0 > > > -----ospfd.conf------ > hostname host1-ospf > password root > enable password root > ! > interface lo > ! > interface eth0 > ! > interface eth1 > ! > interface dummy0 > ! > router ospf > passive-interafec dummy0 > passive-interface lo > network 10.0.0.0/8 area 0 > network 192.168.0.1/32 area 0 > ! > line vty > exec-timeout 0 0 > > -----mplsd.conf----- > hostname host1-mpls > password root > enable password root > ! > mpls ldp > ! > interface lo > ! > interface eth0 > ! > interface eth1 > mpls ldp > ! > interface dummy0 > ! > line vty > exec-timeout 0 0 > > I run mplsd -f /etc/mplds.conf -P 2603, but i don't see any out label > messages. Thanks for any help you can give me. > > ~rsw > > > ------------------------------------------------------- > The SF.Net email is sponsored by EclipseCon 2004 > Premiere Conference on Open Tools Development and Integration > See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. > http://www.eclipsecon.org/osdn > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu jl...@mi... |
|
From: Rudy S W. <rs...@an...> - 2004-02-01 20:03:45
|
Hello, I've setup a point to point connection between two linux red hat 2.4.18 machines (installed mpls/ldp through the rpms). I'm trying to setup up an LDP session between these two machines. Zebra and ospfd are both running and I can see an ospf adjacency (telnet to ospfd, to run show ip ospf neighbor). Then I invoke mplsd on both machines with a mpld.conf file where i setup mpls ldp on eth1 (for both machines). When i log into mplsd and try the show ldp neighbor command, mplsd gets a seg fault.( I can run show ldp session, database, discovery with no problem). This only seems to happen when the other mpls daemon is running. Am I doing something wrong? Here are my .conf files -----zebra.conf------ hostname host1-zebra password root enable password root service advanced-vty ! interface lo ! interface dummy0 ip address 192.168.0.1/32 ! interface eth0 ip address 10.3.3.1/30 ! interface eth1 ip address 10.3.3.5/30 ! line vty exec-timeout 0 0 -----ospfd.conf------ hostname host1-ospf password root enable password root ! interface lo ! interface eth0 ! interface eth1 ! interface dummy0 ! router ospf passive-interafec dummy0 passive-interface lo network 10.0.0.0/8 area 0 network 192.168.0.1/32 area 0 ! line vty exec-timeout 0 0 -----mplsd.conf----- hostname host1-mpls password root enable password root ! mpls ldp ! interface lo ! interface eth0 ! interface eth1 mpls ldp ! interface dummy0 ! line vty exec-timeout 0 0 I run mplsd -f /etc/mplds.conf -P 2603, but i don't see any out label messages. Thanks for any help you can give me. ~rsw |
|
From: Ramon C. <cas...@in...> - 2004-02-01 14:42:39
|
Hi all, For the upcoming 2.0 release for 2.6 kernels, we would welcome (even if it is in very crude form) docs and howtos, or even pointers to RTFM describing this + any mpls - linux specific configs. Of course, only if you feel like... I can take care of formatting them to docbook and put them on the web page with the user and devel guides. Thanks, Ramon On Sun, 1 Feb 2004, Joseph Grzybowski wrote: > I have been able to set up a two node network using VMware. > > On Friday, January 23, 2004 8:57 AM James R. Leu wrote: > > http://user-mode-linux.sf.net/ |
|
From: Joseph G. <jgr...@sn...> - 2004-02-01 13:56:24
|
I have been able to set up a two node network using VMware. > On Friday, January 23, 2004 8:57 AM James R. Leu wrote: > > You should look into using user-mode-linux. With it I'm able to run > a 6 node network on my laptop. > > http://user-mode-linux.sf.net/ > > On Fri, Jan 23, 2004 at 01:08:00AM -0800, Natz V wrote: > > Hi Guys > > > > > > Through the UI, can I check whether its possible to loopback > and have two sessions in the same machine communicating with each > other... > > > > Thank you in Advance > > > > LDP kid > > |
|
From: James R. L. <jl...@da...> - 2004-01-30 16:54:08
|
Check to make sure that the byte and packet counters for the out and in
segments are incrementing. (cat /proc/net/mpls/mpls_*)
Check that you have the correct arp entries (arp -a)
Read the debugging section at 'http://mpls-linux.sf.net/'
On Fri, Jan 30, 2004 at 02:58:22PM +0100, Jung J?nos wrote:
> Hi!
>
> Now i have an lesp between to hosts:
> (thx2 Christian&Ramon)
>
> Host A:
> eth1: 10.120.0.5
> ---------------
> Host B
> eth1: 10.120.0.6
>
> A:
> mplsadm2 -AO 0 (out: key1)
> mplsadm2 -O key1 -o push:gen:64:set:eth1:ipv4:10.120.0.6
> ip route add 10.120.0.6 /32 dev eth1 lsp key1
> mplsadm2 -AI gen:65:0
> mplsadm2 -L eth1:0
> B:
> mplsadm2 -AO 0 (out: key1)
> mplsadm2 -O key1 -o push:gen:65:set:eth1:ipv4:10.120.0.5
> ip route add 10.120.0.5 /32 dev eth1 lsp key1
> mplsadm2 -AI gen:64:0
> mplsadm2 -L eth1:0
> -----------------
> Ping work between them.
> I captured the traffic (ping) with ethereal.
> (ethereal is installed on host B)
> It says:
> The MPLS packets of B are as written in THE BOOK,
> Source Dest. Prot info
> 10.120.0.6 10.120.0.5 ICMP Echo ping req
>
> 102B
> ETHERNET
> MPLS
> IP
> ICMP
>
> But the packets of A are wrong:
> Src dest Prot info
> Tokyo 00:f0:ba 0a:78:00:05:0a:78 0xd873 EthernetII
> (it variates) (always the same)
> WHAT IS THIS? this are not correst MAC addresses..
> (i have intel N.adapters)
>
> it is 102B long
> Ethereal decodes the BoS bit to 0!
>
> But ping works! IP works over MPLS!
> i can establish even an SSH connection!
>
> What is wrong?
> Is my Linux kernel not ok, or is this a fault of Ethereal?
> Or onother one bug of MPLS-linux?
>
> Regards!
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> -------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> mpls-linux-general mailing list
> mpl...@li...
> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
--
James R. Leu
jl...@mi...
|
|
From: <jj...@hs...> - 2004-01-30 14:01:33
|
Hi! Now i have an lesp between to hosts: (thx2 Christian&Ramon) Host A: eth1: 10.120.0.5 --------------- Host B eth1: 10.120.0.6 A: mplsadm2 -AO 0 (out: key1) mplsadm2 -O key1 -o push:gen:64:set:eth1:ipv4:10.120.0.6 ip route add 10.120.0.6 /32 dev eth1 lsp key1 mplsadm2 -AI gen:65:0 mplsadm2 -L eth1:0 B: mplsadm2 -AO 0 (out: key1) mplsadm2 -O key1 -o push:gen:65:set:eth1:ipv4:10.120.0.5 ip route add 10.120.0.5 /32 dev eth1 lsp key1 mplsadm2 -AI gen:64:0 mplsadm2 -L eth1:0 ----------------- Ping work between them. I captured the traffic (ping) with ethereal. (ethereal is installed on host B) It says: The MPLS packets of B are as written in THE BOOK, Source Dest. Prot info 10.120.0.6 10.120.0.5 ICMP Echo ping req 102B ETHERNET MPLS IP ICMP But the packets of A are wrong: Src dest Prot info Tokyo 00:f0:ba 0a:78:00:05:0a:78 0xd873 EthernetII (it variates) (always the same) WHAT IS THIS? this are not correst MAC addresses.. (i have intel N.adapters) it is 102B long Ethereal decodes the BoS bit to 0! But ping works! IP works over MPLS! i can establish even an SSH connection! What is wrong? Is my Linux kernel not ok, or is this a fault of Ethereal? Or onother one bug of MPLS-linux? Regards! |
|
From: Christian <c.t...@gm...> - 2004-01-27 04:16:54
|
=2D----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello! Am Freitag, 23. Januar 2004 15:19 schrieb Jung Janos: > mplsadm2 -AO 0 (out: key1) > mplsadm2 -O key1 -o push:gen:16:set:eth1:ipv4:10.120.0.6 > ip route add 10.120.0.6/32 lsp key1 > #out:RTNETLINK answers: no such device You need to specify your Device like: ip route add 10.120.0.6 /32 dev eth1 lsp key1 I have written a litte script for the setup of a static route perhaps you f= ind it helpful: =2D ---8<--- #!/bin/bash # # Script to setup a static mpls route between two hosts # # December 2003 - Christan Tr=F6ster # IP=3D"1.1.1.1" KEY1=3D`mplsadm2 -A -O 0 2> /dev/null | grep Key | cut -d" " -f2` DEVICE=3D"eth1" echo "Setting up mpls route to $IP via $DEVICE using key $KEY1..." mplsadm2 -O $KEY1 -o push:gen:16:set:$DEVICE:ipv4:$IP ip route add $IP/32 dev $DEVICE lsp $KEY1 mplsadm2 -A -I gen:17:0 mplsadm2 -L $DEVICE:0 echo "done" =2D ---8<--- Bye, Christian =2D -- | mailto:c.t...@gm... | http://www.fen-net.de/~na2929 | =2D----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAETMfCXo2j/oAbHQRAuZzAJ4pYyqKqJhIlBgxcwKIwvPj8KmaTQCfYZZ6 Ce8+HFezqxyajUgCxTi20XE=3D =3DbzoG =2D----END PGP SIGNATURE----- |
|
From: Ramon C. <cas...@in...> - 2004-01-23 15:30:26
|
On Fri, 23 Jan 2004, Jung Janos wrote: > Hi! > > mplsadm2 -AO 0 (out: key1) > mplsadm2 -O key1 -o push:gen:16:set:eth1:ipv4:10.120.0.6 > ip route add 10.120.0.6/32 lsp key1 > #out:RTNETLINK answers: no such device I am not really sure (I am mostly working on the 2.6 version with a new mplsadm and debugging mechanisms that will make it easier to track bugs), but I wonder if the eth1 interfcace is MPLS enabled? Try setting the labelspace (e.g. to 0) before actually mapping the traffic to the LSP mplsadm2 -L eth1:0 before mplsadm2 -O key1 -o push:gen:16:set:eth1:ipv4:10.120.0.6 Idem for host B > what does it mean? what is RTNETLINK? The answer to that question is complicated and out of the scope of this mail :-P but netlink is a linux kernel mechanism that allows user space applications to communicate with parts of the kernel (for example to get/set routes). > i think this host can not send > IP packets over mpls :) It's being lazy.... Not sure if it answers your question, but it is worth trying. Regards, R. |
533 messages has been excluded from this view by a project administrator.
