mpls-linux-general Mailing List for MPLS for Linux (Page 79)
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: Jorge <jo...@sa...> - 2005-03-17 07:36:25
|
Hello, Anybody knows what is "Pim's RSVP-TE implementation". It is operational?, and where i can found it? I think that it is interesting for me. Thanks for your attention Jorge Garcia jo...@sa... |
|
From: James R. L. <jl...@mi...> - 2005-03-16 19:57:57
|
The p4 server is up and running again thanks to the generosity of Perforce Software for providing me and my open source projects a free 15 user open source developers license for their fantastic p4 RCS system. On Tue, Mar 15, 2005 at 05:34:17PM +0800, Timothy\(???x\) wrote: > Hi James, >=20 > The p4 server seems not ready for us to login. >=20 > [root@localhost bin]# ./p4 client > Perforce client error: > Connect to server failed; check $P4PORT. > TCP connect to leu.dangermen.com:1666 failed. > connect: leu.dangermen.com:1666: Connection refused >=20 > Thanks for your help. >=20 > Best Regards >=20 > Timothy > 2005.03.15 --=20 James R. Leu jl...@mi... |
|
From: James R. L. <jl...@mi...> - 2005-03-16 16:38:32
|
On Thu, Mar 17, 2005 at 12:33:22AM +0800, Timothy\(???x\) wrote: > Hi James, >=20 > May I ask how's the process of renewing P4 server going? >=20 > I want to check when I can get the latest code. I submitted the agreement yesterday, now I'm just waiting for them to issue me a license file. I'll let you know when it back up and running. Sorry for the delay. > Thanks. >=20 > Best Regards=20 >=20 > Timothy > 2005.03.17 --=20 James R. Leu jl...@mi... |
|
From: <m9...@cn...> - 2005-03-16 16:27:14
|
Hi James, May I ask how's the process of renewing P4 server going? I want to check when I can get the latest code. Thanks. Best Regards=20 Timothy 2005.03.17 |
|
From: James R. L. <jl...@mi...> - 2005-03-16 15:49:48
|
On Wed, Mar 16, 2005 at 03:13:29PM +1100, Dylan Syme wrote:
> Hi
>=20
> I have been trying to set up LSPs between two computers running=20
> mpls-enabled kernels (FC2). I am able to compile the patched kernel sour=
ce=20
> code OK, boot into the new kernel and I run the following commands:
>=20
> # mpls nhlfe add key 0
> # mpls nhlfe change key 0x2 instructions push gen 32 nexthop eth0 ipv4=20
> 192.168.1.1
Please issue the following commands and send the output to the list:
cat /sys/mpls/*
# issue the command the fails
dmesg | grep MPLS
I prefer is you use the 'script' command to capture the output as opposed
to cutting and pasting, but I'll take the output anyway you can get it.
> The second of these commands returns:
>=20
> RTNETLINK answers: Operation not permitted.
>=20
> The arguments in the command line are consistent with my network setup, a=
nd=20
> all the other basic instructions (eg. ip route add, mpls labelspace, mpls=
=20
> ilm add) all work as expected.
>=20
> Can you help?
>=20
> cheers
>=20
> Dylan Syme
>=20
>=20
>=20
>=20
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick
> _______________________________________________
> 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...> - 2005-03-16 15:41:05
|
On Wed, Mar 16, 2005 at 01:03:39PM +0100, Sasa Pavlek wrote:
> Hy!
> i saw on previous versions that all outgoing/incoming labels, and keys=20
> were stored in few files in /sys/mpls directory.
I deprecated the SYSFS and PROCFS interface to mpls-linux. The mpls
utility is the only way to see what MPLS entities exist.
mpls ilm show
mpls nhlfe show
mpls labelspace show
mpls xc show
>=20
> But in last version i cant find any of those files.. so im interested,=20
> are those files saved only in memory?
>=20
> Ty!
>=20
>=20
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick
> _______________________________________________
> 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...> - 2005-03-16 15:39:14
|
I'm not sure what you mean by your question. There is already a release of quagga-mpls that has the full MPLS infrastructure. The current release allows for creating static LSPs, creating tunnel interfaces and binding routes to a LSP. What it lacks is protocols specific support. In other words BGP cannot install labels yet, LDP does work, and there is not a RSVP-TE implementations. On Wed, Mar 16, 2005 at 11:29:13AM +0100, Sasa Pavlek wrote: > When would you release the new quagga-mpls with the full MPLS=20 > infrastructure. >=20 >=20 > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu jl...@mi... |
|
From: Sasa P. <sas...@fe...> - 2005-03-16 11:59:34
|
Hy! i saw on previous versions that all outgoing/incoming labels, and keys were stored in few files in /sys/mpls directory. But in last version i cant find any of those files.. so im interested, are those files saved only in memory? Ty! |
|
From: Sasa P. <sas...@fe...> - 2005-03-16 10:25:15
|
When would you release the new quagga-mpls with the full MPLS infrastructure. |
|
From: Dylan S. <sy...@ho...> - 2005-03-16 04:14:35
|
Hi I have been trying to set up LSPs between two computers running mpls-enabled kernels (FC2). I am able to compile the patched kernel source code OK, boot into the new kernel and I run the following commands: # mpls nhlfe add key 0 # mpls nhlfe change key 0x2 instructions push gen 32 nexthop eth0 ipv4 192.168.1.1 The second of these commands returns: RTNETLINK answers: Operation not permitted. The arguments in the command line are consistent with my network setup, and all the other basic instructions (eg. ip route add, mpls labelspace, mpls ilm add) all work as expected. Can you help? cheers Dylan Syme |
|
From: James R. L. <jl...@mi...> - 2005-03-15 16:29:32
|
Grrr. My OSS license through Perforce expired today. I'm in the process of renewing it. Hopefully it will only take a couple of hours. On Tue, Mar 15, 2005 at 05:34:17PM +0800, Timothy\(???x\) wrote: > Hi James, >=20 > The p4 server seems not ready for us to login. >=20 > [root@localhost bin]# ./p4 client > Perforce client error: > Connect to server failed; check $P4PORT. > TCP connect to leu.dangermen.com:1666 failed. > connect: leu.dangermen.com:1666: Connection refused >=20 > Thanks for your help. >=20 > Best Regards >=20 > Timothy > 2005.03.15 --=20 James R. Leu jl...@mi... |
|
From: James R. L. <jl...@mi...> - 2005-03-15 15:24:03
|
On Tue, Mar 15, 2005 at 09:17:39AM +0100, Jorge Garc?a wrote: > Hello, > I'm trying to run quagga-mpls, i have created links and in the next > step, ./update-autotools, give me this errors: You need to install the following three packages: autoconf libtool automake There are existing RPMs or deb's for these. Here are the RPM versions I use: autoconf-2.59-5 libtool-1.5.6-4.FC3.1 automake-1.9.5-1 > [root@localhost quagga-mpls]# ./update-autotools This ./update-autotools > script is deprecated, and will be removed at some stage.Please use the > 'autoreconf' command included with autoconf. > TOOLS VERIONS: > ./update-autotools: line 15: autoheader: command not found > ./update-autotools: line 15: autoconf: command not found > ./update-autotools: line 15: libtool: command not found > ./update-autotools: line 15: libtoolize: command not found > ./update-autotools: line 15: aclocal: command not found > ./update-autotools: line 15: automake: command not found > UPDATING: > ./update-autotools: line 19: aclocal: command not found > ./update-autotools: line 20: autoheader: command not found > ./update-autotools: line 21: autoconf: command not found > ./update-autotools: line 22: libtoolize: command not found > ./update-autotools: line 23: automake: command not found >=20 >=20 > What is wrong? >=20 > Thanks for your attention, >=20 > Jorge Garcia > jo...@sa... >=20 >=20 >=20 > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu jl...@mi... |
|
From: Sasa P. <sas...@fe...> - 2005-03-15 13:29:10
|
To improve MPLS performance do this:
1. in /usr/src/linux/include/net/mpls.h file comment following lines:
/* Comment this to suppress MPLS_DEBUG calls */
//#define MPLS_ENABLE_DEBUG 1
/* Comment this to suppress TRACING enter/exit functions */
//#define MPLS_ENABLE_DEBUG_FUNC 1
2. in /usr/src/linux/net/mpls/mpls_utils.c do this:
void
mpls_skb_dump (struct sk_buff* sk)
{
return;
/* THIS IS IRRELEVANT
unsigned int i;
if (!mpls_debug)
return;
printk("MPLS mpls_skb_dump: from %s with len %d (%d)"
"headroom=%d tailroom=%d\n",
sk->dev?sk->dev->name:" net stack ",
sk->len,
sk->truesize,
skb_headroom(sk),
skb_tailroom(sk));
for (i=(unsigned int)sk->head; i<=(unsigned int)sk->tail; i++) {
if (i == (unsigned int)sk->data)
printk("{");
if (i == (unsigned int)sk->h.raw)
printk("#");
if (i == (unsigned int)sk->nh.raw)
printk("|");
if (i == (unsigned int)sk->mac.raw)
printk("*");
printk("%02x",*((unsigned char*)i));
if (i == (unsigned int)sk->tail)
printk("}");
}
printk("\n");
*/
}
With this, I improve throughput 40 times. (over 6 MPLS nodes with
netperf; Source and Destination are IP nodes)
THROUGHPUT BEFORE: 0.35 Mbps
THROUGHPUT AFTER: 13.01 Mbps !!!!
|
|
From: <m9...@cn...> - 2005-03-15 09:28:03
|
Hi James,
The p4 server seems not ready for us to login.
[root@localhost bin]# ./p4 client
Perforce client error:
Connect to server failed; check $P4PORT.
TCP connect to leu.dangermen.com:1666 failed.
connect: leu.dangermen.com:1666: Connection refused
Thanks for your help.
Best Regards
Timothy
2005.03.15 |
|
From: Jorge <jo...@sa...> - 2005-03-15 08:11:57
|
Hello, I'm trying to run quagga-mpls, i have created links and in the next step, ./update-autotools, give me this errors: [root@localhost quagga-mpls]# ./update-autotools This ./update-autotools script is deprecated, and will be removed at some stage.Please use the 'autoreconf' command included with autoconf. TOOLS VERIONS: ./update-autotools: line 15: autoheader: command not found ./update-autotools: line 15: autoconf: command not found ./update-autotools: line 15: libtool: command not found ./update-autotools: line 15: libtoolize: command not found ./update-autotools: line 15: aclocal: command not found ./update-autotools: line 15: automake: command not found UPDATING: ./update-autotools: line 19: aclocal: command not found ./update-autotools: line 20: autoheader: command not found ./update-autotools: line 21: autoconf: command not found ./update-autotools: line 22: libtoolize: command not found ./update-autotools: line 23: automake: command not found What is wrong? Thanks for your attention, Jorge Garcia jo...@sa... |
|
From: James R. L. <jl...@mi...> - 2005-03-15 03:32:35
|
I just submitted code which should fix your problem. I tested it here
and was able to start up zebra and ldpd.
On Sun, Mar 13, 2005 at 03:50:34PM +0800, ?????x wrote:
> Hi James,
>=20
> I find a small bug in quagga-mpls/zebra/mpls_null.c.
> Because my kernel has no mpls enhancement. So I modify "mpls_method =3D m=
pls_netlink.o" to "mpls_method =3D mpls_null.o".
> But when I compile, there was an error.
> Finally I find out there was a "missing part" (shown below) in "quagga-mp=
ls/zebra/mpls_null.c". After I add this part. Then I pass the compile.
> int
> mpls_ctrl_tunnel_register(struct interface *ifp, int update)
>=20
> {
>=20
> return 0;
>=20
> }
>=20
> =20
>=20
> int
>=20
> mpls_ctrl_tunnel_unregister(struct interface *ifp)
>=20
> {
>=20
> return 0;
>=20
> }
>=20
>=20
>=20
> Back to the problem (can't run zebra) I am facing now.
> I try to use GDB. Below is some information from GDB.
>=20
> [root@localhost zebra]# gdb zebra
> GNU gdb Red Hat Linux (6.1post-1.20040607.41rh)
> Copyright 2004 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you =
are
> welcome to change it and/or distribute copies of it under certain conditi=
ons.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB. Type "show warranty" for detail=
s.
> This GDB was configured as "i386-redhat-linux-gnu"...Using host libthread=
_db library "/lib/tls/libthread_db.so.1".
>=20
> (gdb)
>=20
> (gdb) run
> Starting program: /root/tim/testing/newtry/quagga-mpls/zebra/zebra
>=20
> Program received signal SIGSEGV, Segmentation fault.
> 0x0037ada3 in strlen () from /lib/tls/libc.so.6
> (gdb) bt
> #0 0x0037ada3 in strlen () from /lib/tls/libc.so.6
> #1 0x0037aae5 in strdup () from /lib/tls/libc.so.6
> #2 0x08066275 in zstrdup (type=3D0, str=3D0x0) at memory.c:115
> #3 0x08058d94 in mpls_out_segment_register (out=3D0xfee52bb4) at mpls_li=
b.c:219
> #4 0x0805c0f3 in netlink_mpls (snl=3D0xfee52c74, h=3D0x1) at rt_netlink.=
c:1844
> #5 0x0805a8f4 in netlink_parse_info (filter=3D0x805c04e <netlink_mpls>,
> nl=3D0x807f548, a=3D0xfee52c84, size=3D16384) at rt_netlink.c:435
> #6 0x0805c24d in mpls_read () at rt_netlink.c:1929
> #7 0x0804cae7 in main (argc=3D1, argv=3D0xfee56de4) at main.c:341
>=20
>=20
> Thanks
>=20
> Timothy
> 2005.03.13
--=20
James R. Leu
jl...@mi...
|
|
From: James R. L. <jl...@mi...> - 2005-03-15 02:25:30
|
On Tue, Mar 15, 2005 at 09:40:31AM +0800, mu2000 wrote: > On Mon, 14 Mar 2005 13:36:46 -0600, James R. Leu <jl...@mi...> wr= ote: >=20 > > Are you using TCP offload cards? Did you compiling your own kernel > > or did you use an RPM? >=20 > I use RPMs you put in http://mpls-linux.sourceforge.net/: kernel > iproute2 iptables quagga on Fedora Core 2 >=20 > > Try setting the advmss on the routes for you ingress/egress. > Not too sure how to do this, would appreciate some hints and/or pointers. >=20 > > How are you setting up the routes on ingress egress? > eth1 eth0 > ipnet--ulm1 -------ulm2 -- ulm3 -- ulm4 -- ipnet >=20 > ulm1 is ingress (also engress, right?), only setup mpls for eth1, the > interface card connecting to ulm2 (mpls nhlfe/ilm) and routes (mpls > msp: route NET via IP spec_nh 0x8847 KEY) for ip of eth0 interface in > ulm 2, ulm2/ulm3 subnet, ulm3/ulm4 subnet and ipnet >=20 > for the egress ulm4, only setup mpls for the interface connecting to > ulm3, and routes to the other direction (route NET via IP spec_nh > 0x8847 KEY, for ulm2/ulm3 subnet, ulm3/ulm4 subnet and ipnet) >=20 > The thing is, ping to ulm2/ulm3/ulm4 all works (with mpls labels) but > TCP gets stuck after three-way handshake (because of the "bad tcp > checksum a792 (->59c2)!"?). but TCP to the ip net works fine (traffic > in the mpls network all labelled) I'm going to add some numbers to the drawing above and then create the full set of commands that you will need to issue. You should be able to extrapolate from that. 1.1.1.0/24 3.3.1.0/24 3.3.2.0/24 3.3.3.0/24 2.2.2.0/24 .1 .2 .1 .2 .1 .2 .1 .2 .1 .2 e1 e2 e1 e2 e1 e2 e1 e2 e1 e2 L0 L1 L2 L3 L4 |A|--------|ulm1|--------|ulm2|--------|ulm3|--------|ulm4|--------|B| -----1000----X-----1100----X-----1200----> <----2000----X-----2100----X-----2200----- 'X' denotes label swap A ----- ip route add 2.2.2.0/24 via 1.1.1.2 advmss 1400 ulm1 ---- mpls nhlfe add key 0 instructions push gen 1000 nexthop eth1 ipv4 3.3.1.2 (key 0x2) ip route add 2.2.2.0/24 via 3.3.1.2 spec_nh 0x8847 0x2 advmss 1400 mpls labelspace add dev eth1 labelspace 0 mpls ilm add label gen 2000 labelspace 0 ulm2 ---- mpls labelspace add dev eth2 labelspace 0 mpls ilm add label gen 1000 labelspace 0 mpls nhlfe add key 0 instructions push gen 1100 nexthop eth1 ipv4 3.3.2.2 (key 0x2) mpls xc add ilm_label gen 1000 ilm_labelspace 0 nhlfe key 0x2 mpls labelspace add dev eth1 labelspace 0 mpls ilm add label gen 2100 labelspace 0 mpls nhlfe add key 0 instructions push gen 2000 nexthop eth2 ipv4 3.3.1.1 (key 0x3) mpls xc add ilm_label gen 2100 ilm_labelspace 0 nhlfe key 0x3 ulm3 ---- mpls labelspace add dev eth2 labelspace 0 mpls ilm add label gen 1100 labelspace 0 mpls nhlfe add key 0 instructions push gen 1200 nexthop eth1 ipv4 3.3.3.2 (key 0x2) mpls xc add ilm_label gen 1100 ilm_labelspace 0 nhlfe key 0x2 mpls labelspace add dev eth1 labelspace 0 mpls ilm add label gen 2200 labelspace 0 mpls nhlfe add key 0 instructions push gen 2100 nexthop eth2 ipv4 3.3.2.1 (key 0x3) mpls xc add ilm_label gen 2200 ilm_labelspace 0 nhlfe key 0x3 ulm4 ---- mpls labelspace add dev eth2 labelspace 0 mpls ilm add label gen 1200 labelspace 0 mpls nhlfe add key 0 instructions push gen 2200 nexthop eth2 ipv4 3.3.3.1 (key 0x3) ip route add 1.1.1.0/24 via 3.3.3.1 spec_nh 0x8847 0x3 advmss 1400 B ----- ip route add 1.1.1.0/24 via 2.2.2.1 advmss 1400 If you have ethernet cards which can support jumbo frames then you can remove the advmss part and increase the MTU on L1,L2,L3 to 9000 Note that I did not add any routes IP routes to ulm2 or ulm3, and only the 1.1.1.0/24 and 2.2.2.0/24 routes to ulm4 and ulm1 (respectivly). > I'll try to capture the normal traffic and send it the list. Just make sure you don't waste time, I'd like to see that same packet as is moves from L0 to L1 or from L3 to L4. >=20 > Thanks, > morris >=20 >=20 > On Mon, 14 Mar 2005 13:36:46 -0600, James R. Leu <jl...@mi...> wr= ote: >=20 > > On Sun, Mar 13, 2005 at 12:21:17PM +0800, mu w wrote: > > > > > > Thanks for taking the time to reply, James. > >=20 > > Are you using TCP offload cards? Did you compiling your own kernel > > or did you use an RPM? > >=20 > > > Yes I can ssh to ulm3 when MPLS is not enabled. The MTU between ulm2 = and > > > ulm3 are 1500 (showed by `ip link or ip add`) and 1494 shown by `mpls > > > nhlfe show`. The set up commands for mpls: > >=20 > > Try setting the advmss on the routes for you ingress/egress. > > How are you setting up the routes on ingress egress? > >=20 > > > > > > outgoing labels on bother interfaces: mpls nhlfe add key/ mpls nhlfe > > > change key KEY_PREVIOUSLY_GENERATED ... > > > incoming labels on both interfaces: mpls labelspace add dev DEV > > > labelspace 0/mpls ilm add label gen LABEL(match with incoming label) > > > labelspace 0 > > > (Do I need to use x= c? > > > But ping works Okay, all request and reply are mpls enabled -- I see = the > > > mpls bits in ethereal and tcpdump) > >=20 > > Are any of your devices just a LSR, ie it should only see labeled packe= ts? > > If so then yes, otherwise no :-) > >=20 > > > If I use 'telnet ulm3 22', I see the initial response: > > > Connected to 192.168.25.1. > > > Escape character is '^]'. > > > SSH-1.99-OpenSSH_3.6.1p2 > > > > > > I have done some more testing with ethereal and tcpdump. In one of the > > > ethereal outputs, I can see the three-way handshake is complete, > > > then ulm3 three sends a [SYN, FIN, ACK, CWR] to ulm1. The ethereal > > > says "Header length: 16 bytes (bogus, must be at least 20)" shown bel= ow: > >=20 > > Can you capture the same packet on the IP only link and then the > > MPLS link and do a full HEX dump of it and send it to the mailing list? > >=20 > > > [frames 2,3,4, threeway handshake, no problem] > > > Frame 5 (95 bytes on wire, 95 bytes captured) > > > Ethernet II, Src: 00:02:44:6a:ff:f8, Dst: 00:50:04:c3:7a:40 > > > MultiProtocol Label Switching Header > > > Internet Protocol, Src Addr: 192.168.13.2 (192.168.13.2), Dst Addr: > > > 192.168.25.1 (192.168.25.1) > > > Transmission Control Protocol, Src Port: ssh (22), Dst Port: 1139 > > > (1139), Seq: 0 > > > Source port: ssh (22) > > > Destination port: 1139 (1139) > > > Sequence number: 0 > > > Header length: 16 bytes (bogus, must be at least 20) > > > > > > Frame 6 (95 bytes on wire, 95 bytes captured) > > > Ethernet II, Src: 00:02:44:6a:ff:f8, Dst: 00:50:04:c3:7a:40 > > > MultiProtocol Label Switching Header > > > Internet Protocol, Src Addr: 192.168.13.2 (192.168.13.2), Dst Addr: > > > 192.168.25.1 (192.168.25.1) > > > Transmission Control Protocol, Src Port: ssh (22), Dst Port: 1139 > > > (1139), Seq: 0 > > > Source port: ssh (22) > > > Destination port: 1139 (1139) > > > Sequence number: 0 > > > Header length: 16 bytes (bogus, must be at least 20) > > > ... > > > > > > > > > From some tcpdump output, it looks like the checksum sometimes goes b= ad > > > -- "bad tcp checksum a792 (->59c2)!" in the following tcpdump output. > > > > > > Thanks, > > > morris > > > > > > `tcp -xvp -i eth0` output > > > > > > 11:47:32.828060 MPLS (label 25, exp 0, [S], ttl 64) > > > IP (tos 0x0, ttl 64, id 45153, offset 0, flags [DF], proto 6, > > > length: 60) 192.168.25.1.1138 > 192.168.13.1.ssh: S [tcp sum ok] > > > 3559799760:3559799760(0) win 5840 <mss 1460,sackOK,timestamp 89820081 > > > 0,nop,wscale 2> > > > 0x0000: 0001 9140 4500 003c b061 4000 4006 e307 ...@E..<.a@.= @... > > > 0x0010: c0a8 1901 c0a8 0d01 0472 0016 d42e 3bd0 .........r..= ..;. > > > 0x0020: 0000 0000 a002 16d0 e44f 0000 0204 05b4 .........O..= .... > > > 0x0030: 0402 080a 055a 8bb1 0000 0000 0103 0302 .....Z......= .... > > > 11:47:32.828813 MPLS (label 125, exp 0, [S], ttl 64) > > > IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto 6, > > > length: 60) 192.168.13.1.ssh > 192.168.25.1.1138: S [tcp sum ok] > > > 2832248515:2832248515(0) ack 3559799761 win 5792 <mss > > > 1460,sackOK,timestamp 153217841 89820081,nop,wscale 2> > > > 0x0000: 0007 d140 4500 003c 0000 4000 4006 9369 ...@E..<..@.= @..i > > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec3 ...........r= .... > > > 0x0020: d42e 3bd1 a012 16a0 9887 0000 0204 05b4 ..;.........= .... > > > 0x0030: 0402 080a 0921 eb31 055a 8bb1 0103 0302 .....!.1.Z..= .... > > > 11:47:32.829452 MPLS (label 25, exp 0, [S], ttl 64) > > > IP (tos 0x0, ttl 64, id 45155, offset 0, flags [DF], proto 6, > > > length: 52) 192.168.25.1.1138 > 192.168.13.1.ssh: . [tcp sum ok] ack 1 > > > win 1460 <nop,nop,timestamp 89820083 153217841> > > > 0x0000: 0001 9140 4500 0034 b063 4000 4006 e30d ...@E..4.c@.= @... > > > 0x0010: c0a8 1901 c0a8 0d01 0472 0016 d42e 3bd1 .........r..= ..;. > > > 0x0020: a8d0 aec4 8010 05b4 d838 0000 0101 080a .........8..= .... > > > 0x0030: 055a 8bb3 0921 eb31 .Z...!.1 > > > 11:47:32.832443 MPLS (label 125, exp 0, [S], ttl 64) > > > IP (tos 0x0, ttl 64, id 17910, offset 0, flags [DF], proto 6, > > > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FE [bad tcp cksum a= 792 > > > (->59c2)!] 1:34(33) ack 1 win 1448 urg 0 <nop,nop,[bad opt]> > > > 0x0000: 0007 d140 4500 004d 45f6 4000 4006 4d62 ...@E..ME.@.= @.Mb > > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r= .... > > > 0x0020: d42e 3bd1 6371 05a8 a792 0000 0101 080a ..;.cq......= .... > > > 0x0030: 0921 eb34 055a 8bb3 5353 482d 312e 3939 .!.4.Z..SSH-= 1.99 > > > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6= .1p2 > > > 0x0050: 0a > > > 11:47:33.033583 MPLS (label 125, exp 0, [S], ttl 64) > > > IP (tos 0x0, ttl 64, id 17912, offset 0, flags [DF], proto 6, > > > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: SFRW [bad tcp cksum > > > a792 (->59c2)!] 2832248516:2832248549(33) win 1448 urg 0 <nop,nop,[ba= d opt]> > > > 0x0000: 0007 d140 4500 004d 45f8 4000 4006 4d60 ...@E..ME.@.= @.M` > > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r= .... > > > 0x0020: d42e 3bd1 62a7 05a8 a792 0000 0101 080a ..;.b.......= .... > > > 0x0030: 0921 ebfe 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-= 1.99 > > > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6= .1p2 > > > 0x0050: 0a . > > > 11:47:33.435402 MPLS (label 125, exp 0, [S], ttl 64) > > > IP (tos 0x0, ttl 64, id 17914, offset 0, flags [DF], proto 6, > > > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FR [bad tcp cksum a= 792 > > > (->59c2)!] 1:34(33) ack 1 win 1448 <nop,nop,[bad opt]> > > > 0x0000: 0007 d140 4500 004d 45fa 4000 4006 4d5e ...@E..ME.@.= @.M^ > > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r= .... > > > 0x0020: d42e 3bd1 6115 05a8 a792 0000 0101 080a ..;.a.......= .... > > > 0x0030: 0921 ed90 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-= 1.99 > > > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6= .1p2 > > > 0x0050: 0a . > > > 11:47:34.239254 MPLS (label 125, exp 0, [S], ttl 64) > > > IP (tos 0x0, ttl 64, id 17916, offset 0, flags [DF], proto 6, > > > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FWE [bad tcp cksum > > > a792 (->59c2)!] 1:38(37) ack 1 win 1448 urg 0 > > > 0x0000: 0007 d140 4500 004d 45fc 4000 4006 4d5c ...@E..ME.@.= @.M\ > > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r= .... > > > 0x0020: d42e 3bd1 5df1 05a8 a792 0000 0101 080a ..;.].......= .... > > > 0x0030: 0921 f0b4 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-= 1.99 > > > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6= .1p2 > > > 0x0050: 0a . > > > 11:47:35.669679 IP (tos 0xc0, ttl 1, id 58574, offset 0, flags [non= e], > > > proto 89, length: 68) 192.168.25.2 > OSPF-ALL.MCAST.NET: OSPFv2, Hello > > > (1), length: 48 > > > Router-ID: 192.168.25.2, Area 0.0.0.100, Authentication Type: > > > none (0) > > > Options: [External] > > > Hello Timer: 10s, Dead Timer 40s, Mask: 255.255.255.0, Prior= ity: 1 > > > Designated Router 192.168.25.2, Backup Designated Router > > > 192.168.25.1 > > > Neighbor List: > > > 192.168.88.2 > > > 0x0000: 45c0 0044 e4ce 0000 0159 1a23 c0a8 1902 E..D.....Y.#= .... > > > 0x0010: e000 0005 0201 0030 c0a8 1902 0000 0064 .......0....= ...d > > > 0x0020: 568c 0000 0000 0000 0000 0000 ffff ff00 V...........= .... > > > 0x0030: 000a 0201 0000 0028 c0a8 1902 c0a8 1901 .......(....= .... > > > 0x0040: c0a8 5802 > > > > > > > > > > > > > > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > > > From: James R. Leu <jleu@mi...> > > > * Re: help: how to ssh to mpls machine?* > > > <http://sourceforge.net/mailarchive/message.php?msg_id=3D11144914> > > > 2005-03-12 17:43 > > > > > > > > > > > > > > > Can you ssh to the machine in question when MPLS is not enabled > > > in the network? What are the exact commands you issuesd on all of ma= chines > > > in your diagram? What is the MTU of the link between ulm2 and ulm3? > > > > > > On Sun, Mar 13, 2005 at 12:39:22AM +0800, mu w wrote: > > > > (ssh to non-mpls machines via mpls net works) -- one more time > > > > > > > > OK the previous email didn"t get through...please forgive me for > > > reposting. > > > > > > > > Hi James and all... > > > > > > > > I built an mpls network using static label switched path, by using= mpls > > > > nhlfe...and mpls ilm...etc (in 2.6.9-1.6_FC2mpls_1_946). Everything > > > > works fine except I cannot ssh to an mpls machine. > > > > The following figure show my network > > > > > > > > ulm1--ulm2---ulm3----ulm4 > > > > > > > > ulm2 and ulm3 are the mpls network, ulm1 and ulm4 are normal IP hos= ts. > > > > ulm1 can ssh to ulm4 (traffic on ulm2 and ulm3 are labeled) but can= not > > > > ssh to ulm3. ping to ulm3 is okay. > > > > > > > > Ethereal shows the mpls traffic for ssh handshake...SYN and stuff b= ut > > > > the connection was reset after a while. > > > > > > > > Is ssh to ulm3 possible? If yes how to setup? > > > > > > > > I have seen question on ftp and ssh questions in the archive but ca= n"t > > > > find any answers. Any help and pointer are appreciated. > > > > > > > > Also, is there a more detailed manual for the mpls command in (unam= e -a > > > > =3D 2.6.9-1.6_FC2mpls_1_946)? > > > > > > Not written yet, but you can issue "mpls help" and it will give you > > > command line syntax. > > > > > > > > > > > > > > > ------------------------------------------------------- > > > SF email is sponsored by - The IT Product Guide > > > Read honest & candid reviews on hundreds of IT Products from real use= rs. > > > Discover which products truly live up to the hype. Start reading now. > > > http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick > > > _______________________________________________ > > > 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 >=20 > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu jl...@mi... |
|
From: mu2000 <mu...@gm...> - 2005-03-15 01:40:43
|
On Mon, 14 Mar 2005 13:36:46 -0600, James R. Leu <jl...@mi...> wrote: > Are you using TCP offload cards? Did you compiling your own kernel > or did you use an RPM? I use RPMs you put in http://mpls-linux.sourceforge.net/: kernel iproute2 iptables quagga on Fedora Core 2 > Try setting the advmss on the routes for you ingress/egress. Not too sure how to do this, would appreciate some hints and/or pointers. > How are you setting up the routes on ingress egress? eth1 eth0 ipnet--ulm1 -------ulm2 -- ulm3 -- ulm4 -- ipnet ulm1 is ingress (also engress, right?), only setup mpls for eth1, the interface card connecting to ulm2 (mpls nhlfe/ilm) and routes (mpls msp: route NET via IP spec_nh 0x8847 KEY) for ip of eth0 interface in ulm 2, ulm2/ulm3 subnet, ulm3/ulm4 subnet and ipnet for the egress ulm4, only setup mpls for the interface connecting to ulm3, and routes to the other direction (route NET via IP spec_nh 0x8847 KEY, for ulm2/ulm3 subnet, ulm3/ulm4 subnet and ipnet) The thing is, ping to ulm2/ulm3/ulm4 all works (with mpls labels) but TCP gets stuck after three-way handshake (because of the "bad tcp checksum a792 (->59c2)!"?). but TCP to the ip net works fine (traffic in the mpls network all labelled) I'll try to capture the normal traffic and send it the list. Thanks, morris On Mon, 14 Mar 2005 13:36:46 -0600, James R. Leu <jl...@mi...> wrote: > On Sun, Mar 13, 2005 at 12:21:17PM +0800, mu w wrote: > > > > Thanks for taking the time to reply, James. > > Are you using TCP offload cards? Did you compiling your own kernel > or did you use an RPM? > > > Yes I can ssh to ulm3 when MPLS is not enabled. The MTU between ulm2 and > > ulm3 are 1500 (showed by `ip link or ip add`) and 1494 shown by `mpls > > nhlfe show`. The set up commands for mpls: > > Try setting the advmss on the routes for you ingress/egress. > How are you setting up the routes on ingress egress? > > > > > outgoing labels on bother interfaces: mpls nhlfe add key/ mpls nhlfe > > change key KEY_PREVIOUSLY_GENERATED ... > > incoming labels on both interfaces: mpls labelspace add dev DEV > > labelspace 0/mpls ilm add label gen LABEL(match with incoming label) > > labelspace 0 > > (Do I need to use xc? > > But ping works Okay, all request and reply are mpls enabled -- I see the > > mpls bits in ethereal and tcpdump) > > Are any of your devices just a LSR, ie it should only see labeled packets? > If so then yes, otherwise no :-) > > > If I use 'telnet ulm3 22', I see the initial response: > > Connected to 192.168.25.1. > > Escape character is '^]'. > > SSH-1.99-OpenSSH_3.6.1p2 > > > > I have done some more testing with ethereal and tcpdump. In one of the > > ethereal outputs, I can see the three-way handshake is complete, > > then ulm3 three sends a [SYN, FIN, ACK, CWR] to ulm1. The ethereal > > says "Header length: 16 bytes (bogus, must be at least 20)" shown below: > > Can you capture the same packet on the IP only link and then the > MPLS link and do a full HEX dump of it and send it to the mailing list? > > > [frames 2,3,4, threeway handshake, no problem] > > Frame 5 (95 bytes on wire, 95 bytes captured) > > Ethernet II, Src: 00:02:44:6a:ff:f8, Dst: 00:50:04:c3:7a:40 > > MultiProtocol Label Switching Header > > Internet Protocol, Src Addr: 192.168.13.2 (192.168.13.2), Dst Addr: > > 192.168.25.1 (192.168.25.1) > > Transmission Control Protocol, Src Port: ssh (22), Dst Port: 1139 > > (1139), Seq: 0 > > Source port: ssh (22) > > Destination port: 1139 (1139) > > Sequence number: 0 > > Header length: 16 bytes (bogus, must be at least 20) > > > > Frame 6 (95 bytes on wire, 95 bytes captured) > > Ethernet II, Src: 00:02:44:6a:ff:f8, Dst: 00:50:04:c3:7a:40 > > MultiProtocol Label Switching Header > > Internet Protocol, Src Addr: 192.168.13.2 (192.168.13.2), Dst Addr: > > 192.168.25.1 (192.168.25.1) > > Transmission Control Protocol, Src Port: ssh (22), Dst Port: 1139 > > (1139), Seq: 0 > > Source port: ssh (22) > > Destination port: 1139 (1139) > > Sequence number: 0 > > Header length: 16 bytes (bogus, must be at least 20) > > ... > > > > > > From some tcpdump output, it looks like the checksum sometimes goes bad > > -- "bad tcp checksum a792 (->59c2)!" in the following tcpdump output. > > > > Thanks, > > morris > > > > `tcp -xvp -i eth0` output > > > > 11:47:32.828060 MPLS (label 25, exp 0, [S], ttl 64) > > IP (tos 0x0, ttl 64, id 45153, offset 0, flags [DF], proto 6, > > length: 60) 192.168.25.1.1138 > 192.168.13.1.ssh: S [tcp sum ok] > > 3559799760:3559799760(0) win 5840 <mss 1460,sackOK,timestamp 89820081 > > 0,nop,wscale 2> > > 0x0000: 0001 9140 4500 003c b061 4000 4006 e307 ...@E..<.a@.@... > > 0x0010: c0a8 1901 c0a8 0d01 0472 0016 d42e 3bd0 .........r....;. > > 0x0020: 0000 0000 a002 16d0 e44f 0000 0204 05b4 .........O...... > > 0x0030: 0402 080a 055a 8bb1 0000 0000 0103 0302 .....Z.......... > > 11:47:32.828813 MPLS (label 125, exp 0, [S], ttl 64) > > IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto 6, > > length: 60) 192.168.13.1.ssh > 192.168.25.1.1138: S [tcp sum ok] > > 2832248515:2832248515(0) ack 3559799761 win 5792 <mss > > 1460,sackOK,timestamp 153217841 89820081,nop,wscale 2> > > 0x0000: 0007 d140 4500 003c 0000 4000 4006 9369 ...@E..<..@.@..i > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec3 ...........r.... > > 0x0020: d42e 3bd1 a012 16a0 9887 0000 0204 05b4 ..;............. > > 0x0030: 0402 080a 0921 eb31 055a 8bb1 0103 0302 .....!.1.Z...... > > 11:47:32.829452 MPLS (label 25, exp 0, [S], ttl 64) > > IP (tos 0x0, ttl 64, id 45155, offset 0, flags [DF], proto 6, > > length: 52) 192.168.25.1.1138 > 192.168.13.1.ssh: . [tcp sum ok] ack 1 > > win 1460 <nop,nop,timestamp 89820083 153217841> > > 0x0000: 0001 9140 4500 0034 b063 4000 4006 e30d ...@E..4.c@.@... > > 0x0010: c0a8 1901 c0a8 0d01 0472 0016 d42e 3bd1 .........r....;. > > 0x0020: a8d0 aec4 8010 05b4 d838 0000 0101 080a .........8...... > > 0x0030: 055a 8bb3 0921 eb31 .Z...!.1 > > 11:47:32.832443 MPLS (label 125, exp 0, [S], ttl 64) > > IP (tos 0x0, ttl 64, id 17910, offset 0, flags [DF], proto 6, > > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FE [bad tcp cksum a792 > > (->59c2)!] 1:34(33) ack 1 win 1448 urg 0 <nop,nop,[bad opt]> > > 0x0000: 0007 d140 4500 004d 45f6 4000 4006 4d62 ...@E..ME.@.@.Mb > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r.... > > 0x0020: d42e 3bd1 6371 05a8 a792 0000 0101 080a ..;.cq.......... > > 0x0030: 0921 eb34 055a 8bb3 5353 482d 312e 3939 .!.4.Z..SSH-1.99 > > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6.1p2 > > 0x0050: 0a > > 11:47:33.033583 MPLS (label 125, exp 0, [S], ttl 64) > > IP (tos 0x0, ttl 64, id 17912, offset 0, flags [DF], proto 6, > > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: SFRW [bad tcp cksum > > a792 (->59c2)!] 2832248516:2832248549(33) win 1448 urg 0 <nop,nop,[bad opt]> > > 0x0000: 0007 d140 4500 004d 45f8 4000 4006 4d60 ...@E..ME.@.@.M` > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r.... > > 0x0020: d42e 3bd1 62a7 05a8 a792 0000 0101 080a ..;.b........... > > 0x0030: 0921 ebfe 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-1.99 > > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6.1p2 > > 0x0050: 0a . > > 11:47:33.435402 MPLS (label 125, exp 0, [S], ttl 64) > > IP (tos 0x0, ttl 64, id 17914, offset 0, flags [DF], proto 6, > > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FR [bad tcp cksum a792 > > (->59c2)!] 1:34(33) ack 1 win 1448 <nop,nop,[bad opt]> > > 0x0000: 0007 d140 4500 004d 45fa 4000 4006 4d5e ...@E..ME.@.@.M^ > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r.... > > 0x0020: d42e 3bd1 6115 05a8 a792 0000 0101 080a ..;.a........... > > 0x0030: 0921 ed90 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-1.99 > > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6.1p2 > > 0x0050: 0a . > > 11:47:34.239254 MPLS (label 125, exp 0, [S], ttl 64) > > IP (tos 0x0, ttl 64, id 17916, offset 0, flags [DF], proto 6, > > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FWE [bad tcp cksum > > a792 (->59c2)!] 1:38(37) ack 1 win 1448 urg 0 > > 0x0000: 0007 d140 4500 004d 45fc 4000 4006 4d5c ...@E..ME.@.@.M\ > > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r.... > > 0x0020: d42e 3bd1 5df1 05a8 a792 0000 0101 080a ..;.]........... > > 0x0030: 0921 f0b4 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-1.99 > > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6.1p2 > > 0x0050: 0a . > > 11:47:35.669679 IP (tos 0xc0, ttl 1, id 58574, offset 0, flags [none], > > proto 89, length: 68) 192.168.25.2 > OSPF-ALL.MCAST.NET: OSPFv2, Hello > > (1), length: 48 > > Router-ID: 192.168.25.2, Area 0.0.0.100, Authentication Type: > > none (0) > > Options: [External] > > Hello Timer: 10s, Dead Timer 40s, Mask: 255.255.255.0, Priority: 1 > > Designated Router 192.168.25.2, Backup Designated Router > > 192.168.25.1 > > Neighbor List: > > 192.168.88.2 > > 0x0000: 45c0 0044 e4ce 0000 0159 1a23 c0a8 1902 E..D.....Y.#.... > > 0x0010: e000 0005 0201 0030 c0a8 1902 0000 0064 .......0.......d > > 0x0020: 568c 0000 0000 0000 0000 0000 ffff ff00 V............... > > 0x0030: 000a 0201 0000 0028 c0a8 1902 c0a8 1901 .......(........ > > 0x0040: c0a8 5802 > > > > > > > > > > ======================================= > > From: James R. Leu <jleu@mi...> > > * Re: help: how to ssh to mpls machine?* > > <http://sourceforge.net/mailarchive/message.php?msg_id=11144914> > > 2005-03-12 17:43 > > > > > > > > > > Can you ssh to the machine in question when MPLS is not enabled > > in the network? What are the exact commands you issuesd on all of machines > > in your diagram? What is the MTU of the link between ulm2 and ulm3? > > > > On Sun, Mar 13, 2005 at 12:39:22AM +0800, mu w wrote: > > > (ssh to non-mpls machines via mpls net works) -- one more time > > > > > > OK the previous email didn"t get through...please forgive me for > > reposting. > > > > > > Hi James and all... > > > > > > I built an mpls network using static label switched path, by using mpls > > > nhlfe...and mpls ilm...etc (in 2.6.9-1.6_FC2mpls_1_946). Everything > > > works fine except I cannot ssh to an mpls machine. > > > The following figure show my network > > > > > > ulm1--ulm2---ulm3----ulm4 > > > > > > ulm2 and ulm3 are the mpls network, ulm1 and ulm4 are normal IP hosts. > > > ulm1 can ssh to ulm4 (traffic on ulm2 and ulm3 are labeled) but cannot > > > ssh to ulm3. ping to ulm3 is okay. > > > > > > Ethereal shows the mpls traffic for ssh handshake...SYN and stuff but > > > the connection was reset after a while. > > > > > > Is ssh to ulm3 possible? If yes how to setup? > > > > > > I have seen question on ftp and ssh questions in the archive but can"t > > > find any answers. Any help and pointer are appreciated. > > > > > > Also, is there a more detailed manual for the mpls command in (uname -a > > > = 2.6.9-1.6_FC2mpls_1_946)? > > > > Not written yet, but you can issue "mpls help" and it will give you > > command line syntax. > > > > > > > > > > ------------------------------------------------------- > > SF email is sponsored by - The IT Product Guide > > Read honest & candid reviews on hundreds of IT Products from real users. > > Discover which products truly live up to the hype. Start reading now. > > http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click > > _______________________________________________ > > mpls-linux-general mailing list > > mpl...@li... > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > -- > James R. Leu > jl...@mi... > > > |
|
From: James R. L. <jl...@mi...> - 2005-03-14 19:47:46
|
On Mon, Mar 14, 2005 at 12:17:02PM +0100, sasa pavlek wrote: > May be I didn't be enough specific. I was monitoring MPLS > traffic on every LSR and TTL is decrementing every MPLS hop. Yes the MPLS TTL _is_ decremented at every hop that is the whole point of i= t. If your refering to the IP TTL being changed at every hop, then your network is configured wrong, and you did not cross connect the ILM and NHLFE, in other words you created a bunch of single hop LSPs that result in L3 lookups at every hop. How are you creating the LSPs? Did you try using quagga-mpls? It has a CLI for creating static LSPs that might be easier for you to use. > On Mon, 14 Mar 2005, Christophe Fillot wrote: >=20 > > On Mon, 14 Mar 2005, Sasa Pavlek wrote: > > > > > One more problem: Why IP TTL is decrementing when transmitting thruog= h MPLS > > > network. I think that IP TTL must be unchanged in MPLS network. > > > > At ingress, the IP TTL is copied into MPLS TTL. At egress, the MPLS TTL > > is copied into the IP TTL. > > > > -- > > Christophe Fillot (cf...@ut...) | Universite de Technologie de Compiegne > > Tel: (+33) 03.44.23.79.02 | Service Informatique - Ingenieur Reseaux > > GSM: (+33) 06.70.50.24.55 | Centre de Recherche de Royallieu > > Fax: (+33) 03.44.23.46.77 | BP 20.529, 60205 Compiegne Cedex > > > > >=20 >=20 >=20 > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick > _______________________________________________ > 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...> - 2005-03-14 19:36:01
|
On Sun, Mar 13, 2005 at 12:21:17PM +0800, mu w wrote: >=20 > Thanks for taking the time to reply, James. Are you using TCP offload cards? Did you compiling your own kernel or did you use an RPM? > Yes I can ssh to ulm3 when MPLS is not enabled. The MTU between ulm2 and= =20 > ulm3 are 1500 (showed by `ip link or ip add`) and 1494 shown by `mpls=20 > nhlfe show`. The set up commands for mpls: Try setting the advmss on the routes for you ingress/egress. How are you setting up the routes on ingress egress? >=20 > outgoing labels on bother interfaces: mpls nhlfe add key/ mpls nhlfe=20 > change key KEY_PREVIOUSLY_GENERATED ... > incoming labels on both interfaces: mpls labelspace add dev DEV=20 > labelspace 0/mpls ilm add label gen LABEL(match with incoming label)=20 > labelspace 0 > (Do I need to use xc?= =20 > But ping works Okay, all request and reply are mpls enabled -- I see the= =20 > mpls bits in ethereal and tcpdump) Are any of your devices just a LSR, ie it should only see labeled packets? If so then yes, otherwise no :-) > If I use 'telnet ulm3 22', I see the initial response: > Connected to 192.168.25.1. > Escape character is '^]'. > SSH-1.99-OpenSSH_3.6.1p2 >=20 > I have done some more testing with ethereal and tcpdump. In one of the=20 > ethereal outputs, I can see the three-way handshake is complete, =20 > then ulm3 three sends a [SYN, FIN, ACK, CWR] to ulm1. The ethereal=20 > says "Header length: 16 bytes (bogus, must be at least 20)" shown below: Can you capture the same packet on the IP only link and then the MPLS link and do a full HEX dump of it and send it to the mailing list? > [frames 2,3,4, threeway handshake, no problem] > Frame 5 (95 bytes on wire, 95 bytes captured) > Ethernet II, Src: 00:02:44:6a:ff:f8, Dst: 00:50:04:c3:7a:40 > MultiProtocol Label Switching Header > Internet Protocol, Src Addr: 192.168.13.2 (192.168.13.2), Dst Addr:=20 > 192.168.25.1 (192.168.25.1) > Transmission Control Protocol, Src Port: ssh (22), Dst Port: 1139=20 > (1139), Seq: 0 > Source port: ssh (22) > Destination port: 1139 (1139) > Sequence number: 0 > Header length: 16 bytes (bogus, must be at least 20) >=20 > Frame 6 (95 bytes on wire, 95 bytes captured) > Ethernet II, Src: 00:02:44:6a:ff:f8, Dst: 00:50:04:c3:7a:40 > MultiProtocol Label Switching Header > Internet Protocol, Src Addr: 192.168.13.2 (192.168.13.2), Dst Addr:=20 > 192.168.25.1 (192.168.25.1) > Transmission Control Protocol, Src Port: ssh (22), Dst Port: 1139=20 > (1139), Seq: 0 > Source port: ssh (22) > Destination port: 1139 (1139) > Sequence number: 0 > Header length: 16 bytes (bogus, must be at least 20) > ... >=20 >=20 > From some tcpdump output, it looks like the checksum sometimes goes bad= =20 > -- "bad tcp checksum a792 (->59c2)!" in the following tcpdump output. >=20 > Thanks, > morris >=20 > `tcp -xvp -i eth0` output >=20 > 11:47:32.828060 MPLS (label 25, exp 0, [S], ttl 64) > IP (tos 0x0, ttl 64, id 45153, offset 0, flags [DF], proto 6,=20 > length: 60) 192.168.25.1.1138 > 192.168.13.1.ssh: S [tcp sum ok]=20 > 3559799760:3559799760(0) win 5840 <mss 1460,sackOK,timestamp 89820081=20 > 0,nop,wscale 2> > 0x0000: 0001 9140 4500 003c b061 4000 4006 e307 ...@E..<.a@.@... > 0x0010: c0a8 1901 c0a8 0d01 0472 0016 d42e 3bd0 .........r....;. > 0x0020: 0000 0000 a002 16d0 e44f 0000 0204 05b4 .........O...... > 0x0030: 0402 080a 055a 8bb1 0000 0000 0103 0302 .....Z.......... > 11:47:32.828813 MPLS (label 125, exp 0, [S], ttl 64) > IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto 6,=20 > length: 60) 192.168.13.1.ssh > 192.168.25.1.1138: S [tcp sum ok]=20 > 2832248515:2832248515(0) ack 3559799761 win 5792 <mss=20 > 1460,sackOK,timestamp 153217841 89820081,nop,wscale 2> > 0x0000: 0007 d140 4500 003c 0000 4000 4006 9369 ...@E..<..@.@..i > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec3 ...........r.... > 0x0020: d42e 3bd1 a012 16a0 9887 0000 0204 05b4 ..;............. > 0x0030: 0402 080a 0921 eb31 055a 8bb1 0103 0302 .....!.1.Z...... > 11:47:32.829452 MPLS (label 25, exp 0, [S], ttl 64) > IP (tos 0x0, ttl 64, id 45155, offset 0, flags [DF], proto 6,=20 > length: 52) 192.168.25.1.1138 > 192.168.13.1.ssh: . [tcp sum ok] ack 1=20 > win 1460 <nop,nop,timestamp 89820083 153217841> > 0x0000: 0001 9140 4500 0034 b063 4000 4006 e30d ...@E..4.c@.@... > 0x0010: c0a8 1901 c0a8 0d01 0472 0016 d42e 3bd1 .........r....;. > 0x0020: a8d0 aec4 8010 05b4 d838 0000 0101 080a .........8...... > 0x0030: 055a 8bb3 0921 eb31 .Z...!.1 > 11:47:32.832443 MPLS (label 125, exp 0, [S], ttl 64) > IP (tos 0x0, ttl 64, id 17910, offset 0, flags [DF], proto 6,=20 > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FE [bad tcp cksum a792= =20 > (->59c2)!] 1:34(33) ack 1 win 1448 urg 0 <nop,nop,[bad opt]> > 0x0000: 0007 d140 4500 004d 45f6 4000 4006 4d62 ...@E..ME.@.@.Mb > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r.... > 0x0020: d42e 3bd1 6371 05a8 a792 0000 0101 080a ..;.cq.......... > 0x0030: 0921 eb34 055a 8bb3 5353 482d 312e 3939 .!.4.Z..SSH-1.99 > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6.1p2 > 0x0050: 0a > 11:47:33.033583 MPLS (label 125, exp 0, [S], ttl 64) > IP (tos 0x0, ttl 64, id 17912, offset 0, flags [DF], proto 6,=20 > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: SFRW [bad tcp cksum=20 > a792 (->59c2)!] 2832248516:2832248549(33) win 1448 urg 0 <nop,nop,[bad op= t]> > 0x0000: 0007 d140 4500 004d 45f8 4000 4006 4d60 ...@E..ME.@.@.M` > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r.... > 0x0020: d42e 3bd1 62a7 05a8 a792 0000 0101 080a ..;.b........... > 0x0030: 0921 ebfe 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-1.99 > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6.1p2 > 0x0050: 0a . > 11:47:33.435402 MPLS (label 125, exp 0, [S], ttl 64) > IP (tos 0x0, ttl 64, id 17914, offset 0, flags [DF], proto 6,=20 > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FR [bad tcp cksum a792= =20 > (->59c2)!] 1:34(33) ack 1 win 1448 <nop,nop,[bad opt]> > 0x0000: 0007 d140 4500 004d 45fa 4000 4006 4d5e ...@E..ME.@.@.M^ > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r.... > 0x0020: d42e 3bd1 6115 05a8 a792 0000 0101 080a ..;.a........... > 0x0030: 0921 ed90 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-1.99 > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6.1p2 > 0x0050: 0a . > 11:47:34.239254 MPLS (label 125, exp 0, [S], ttl 64) > IP (tos 0x0, ttl 64, id 17916, offset 0, flags [DF], proto 6,=20 > length: 77) 192.168.13.1.ssh > 192.168.25.1.1138: FWE [bad tcp cksum=20 > a792 (->59c2)!] 1:38(37) ack 1 win 1448 urg 0 > 0x0000: 0007 d140 4500 004d 45fc 4000 4006 4d5c ...@E..ME.@.@.M\ > 0x0010: c0a8 0d01 c0a8 1901 0016 0472 a8d0 aec4 ...........r.... > 0x0020: d42e 3bd1 5df1 05a8 a792 0000 0101 080a ..;.]........... > 0x0030: 0921 f0b4 055a 8bb3 5353 482d 312e 3939 .!...Z..SSH-1.99 > 0x0040: 2d4f 7065 6e53 5348 5f33 2e36 2e31 7032 -OpenSSH_3.6.1p2 > 0x0050: 0a . > 11:47:35.669679 IP (tos 0xc0, ttl 1, id 58574, offset 0, flags [none],= =20 > proto 89, length: 68) 192.168.25.2 > OSPF-ALL.MCAST.NET: OSPFv2, Hello=20 > (1), length: 48 > Router-ID: 192.168.25.2, Area 0.0.0.100, Authentication Type:=20 > none (0) > Options: [External] > Hello Timer: 10s, Dead Timer 40s, Mask: 255.255.255.0, Priority:= 1 > Designated Router 192.168.25.2, Backup Designated Router=20 > 192.168.25.1 > Neighbor List: > 192.168.88.2 > 0x0000: 45c0 0044 e4ce 0000 0159 1a23 c0a8 1902 E..D.....Y.#.... > 0x0010: e000 0005 0201 0030 c0a8 1902 0000 0064 .......0.......d > 0x0020: 568c 0000 0000 0000 0000 0000 ffff ff00 V............... > 0x0030: 000a 0201 0000 0028 c0a8 1902 c0a8 1901 .......(........ > 0x0040: c0a8 5802 =20 >=20 >=20 >=20 >=20 > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > From: James R. Leu <jleu@mi...> > * Re: help: how to ssh to mpls machine?*=20 > <http://sourceforge.net/mailarchive/message.php?msg_id=3D11144914> =20 > 2005-03-12 17:43 >=20 >=20 >=20 >=20 > Can you ssh to the machine in question when MPLS is not enabled > in the network? What are the exact commands you issuesd on all of machin= es > in your diagram? What is the MTU of the link between ulm2 and ulm3? >=20 > On Sun, Mar 13, 2005 at 12:39:22AM +0800, mu w wrote: > > (ssh to non-mpls machines via mpls net works) -- one more time > >=20 > > OK the previous email didn"t get through...please forgive me for=20 > reposting. > >=20 > > Hi James and all... > >=20 > > I built an mpls network using static label switched path, by using mpl= s=20 > > nhlfe...and mpls ilm...etc (in 2.6.9-1.6_FC2mpls_1_946). Everything=20 > > works fine except I cannot ssh to an mpls machine. > > The following figure show my network > > =20 > > ulm1--ulm2---ulm3----ulm4=20 > > =20 > > ulm2 and ulm3 are the mpls network, ulm1 and ulm4 are normal IP hosts.= =20 > > ulm1 can ssh to ulm4 (traffic on ulm2 and ulm3 are labeled) but cannot= =20 > > ssh to ulm3. ping to ulm3 is okay. > >=20 > > Ethereal shows the mpls traffic for ssh handshake...SYN and stuff but= =20 > > the connection was reset after a while. > >=20 > > Is ssh to ulm3 possible? If yes how to setup? > >=20 > > I have seen question on ftp and ssh questions in the archive but can"t= =20 > > find any answers. Any help and pointer are appreciated. > >=20 > > Also, is there a more detailed manual for the mpls command in (uname -a= =20 > > =3D 2.6.9-1.6_FC2mpls_1_946)? =20 >=20 > Not written yet, but you can issue "mpls help" and it will give you > command line syntax. >=20 >=20 >=20 >=20 > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick > _______________________________________________ > 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...> - 2005-03-14 19:12:49
|
On Mon, Mar 14, 2005 at 10:56:46AM +0100, Sasa Pavlek wrote: > Did anyone tested MPLS network with netperf, netpipe, ping or some other= =20 > network tool, becouse I had some problems with measuring MPLS throughput= =20 > and round-trip-time. mpls-linux is not optimized for throughput it is optimized for flexibility and correctness, you would have seen that had you searched the mailing list archives. I would be happy to work with someone if they would like to try and optimize mpls-linux for throughput. > Network: >=20 > Source Ingress = =20 > Ingress Destination > Egress =20 > Egress =20 > =20 > --------- --------- --------- --------- =20 > --------- --------- --------- > | ip =20 > |--------|ip/mpls|--------|ip/mpls|--------|ip/mpls|--------|ip/mpls|----= ----|ip/mpls|-------=20 > | ip | > | node | | node 1| | node 2| | node 3| |=20 > node 4| | node 5| | node | > --------- --------- --------- --------- =20 > --------- --------- --------- > | | =20 > | > | | =20 > | > | | =20 > | > --------- --------- =20 > --------- > =20 > |ip/mpls|-------------------------|ip/mpls|-------------------------|ip/m= pls| > | node 6| | node 7| =20 > | node 8| > --------- --------- =20 > --------- >=20 > All nodes are connected with 100 Mbps links. >=20 > I try to compare IP and MPLS throughput and RTT, but I got odd results.= =20 > For IP throughput between Source and Destination with netperf, I got=20 > approximately 90 Mbps over 1,2,3,4 and 5 mpls node, but for MPLS=20 > throughput with same tool I got only 350 kbps. What are the MTU's of your links? Did you turn off debugging? Did you recompile with MPLS_DEBUG turned off? > When I ping (ping -qfc 10000 -s 8 dest_ip) Destination from Source, for= =20 > MPLS I got 100 times bigger min/avg/max RTT resutls then for IP. But=20 > for MPLS is faster total ping time then for IP. >=20 > One more problem: Why IP TTL is decrementing when transmitting thruogh=20 > MPLS network. I think that IP TTL must be unchanged in MPLS network. Wrong. Read RFC3031. Is is a user-configurable option. In mpls-linux it is configurable via the 'propagate_ttl' NHLFE option. >=20 > =09 > --------- --------- --------- --------- --------- --------- --------- > | ip | ------|ip/mpls|-------|ip/mpls|-------|ip/mpls|-------|ip/mpls|= -------|ip/mpls|-------| ip | > | node 1| | node 1| | node 2| | node 3| | node 4| | node 5| | node 2| > --------- --------- --------- --------- --------- --------- --------- > | | | > | | | > | | | > --------- --------- --------- > |ip/mpls|-----------------------|ip/mpls|-----------------------|ip/mpl= s| > | node 6| | node 7| | node 8| > --------- --------- --------- PS please make you ascii drawings fit an 80 col terminal. --=20 James R. Leu jl...@mi... |
|
From: sasa p. <sp...@pi...> - 2005-03-14 11:17:13
|
May be I didn't be enough specific. I was monitoring MPLS traffic on every LSR and TTL is decrementing every MPLS hop. On Mon, 14 Mar 2005, Christophe Fillot wrote: > On Mon, 14 Mar 2005, Sasa Pavlek wrote: > > > One more problem: Why IP TTL is decrementing when transmitting thruogh MPLS > > network. I think that IP TTL must be unchanged in MPLS network. > > At ingress, the IP TTL is copied into MPLS TTL. At egress, the MPLS TTL > is copied into the IP TTL. > > -- > Christophe Fillot (cf...@ut...) | Universite de Technologie de Compiegne > Tel: (+33) 03.44.23.79.02 | Service Informatique - Ingenieur Reseaux > GSM: (+33) 06.70.50.24.55 | Centre de Recherche de Royallieu > Fax: (+33) 03.44.23.46.77 | BP 20.529, 60205 Compiegne Cedex > > |
|
From: Christophe F. <cf...@ut...> - 2005-03-14 10:09:14
|
On Mon, 14 Mar 2005, Sasa Pavlek wrote: > One more problem: Why IP TTL is decrementing when transmitting thruogh MPLS > network. I think that IP TTL must be unchanged in MPLS network. At ingress, the IP TTL is copied into MPLS TTL. At egress, the MPLS TTL is copied into the IP TTL. -- Christophe Fillot (cf...@ut...) | Universite de Technologie de Compiegne Tel: (+33) 03.44.23.79.02 | Service Informatique - Ingenieur Reseaux GSM: (+33) 06.70.50.24.55 | Centre de Recherche de Royallieu Fax: (+33) 03.44.23.46.77 | BP 20.529, 60205 Compiegne Cedex |
|
From: Sasa P. <sas...@fe...> - 2005-03-14 09:52:46
|
Did anyone tested MPLS network with netperf, netpipe, ping or some other
network tool, becouse I had some problems with measuring MPLS throughput
and round-trip-time.
Network:
Source Ingress
Ingress Destination
Egress
Egress
--------- --------- --------- ---------
--------- --------- ---------
| ip
|--------|ip/mpls|--------|ip/mpls|--------|ip/mpls|--------|ip/mpls|--------|ip/mpls|-------
| ip |
| node | | node 1| | node 2| | node 3| |
node 4| | node 5| | node |
--------- --------- --------- ---------
--------- --------- ---------
| |
|
| |
|
| |
|
--------- ---------
---------
|ip/mpls|-------------------------|ip/mpls|-------------------------|ip/mpls|
| node 6| | node 7|
| node 8|
--------- ---------
---------
All nodes are connected with 100 Mbps links.
I try to compare IP and MPLS throughput and RTT, but I got odd results.
For IP throughput between Source and Destination with netperf, I got
approximately 90 Mbps over 1,2,3,4 and 5 mpls node, but for MPLS
throughput with same tool I got only 350 kbps.
When I ping (ping -qfc 10000 -s 8 dest_ip) Destination from Source, for
MPLS I got 100 times bigger min/avg/max RTT resutls then for IP. But
for MPLS is faster total ping time then for IP.
One more problem: Why IP TTL is decrementing when transmitting thruogh
MPLS network. I think that IP TTL must be unchanged in MPLS network.
|
|
From: James R. L. <jl...@mi...> - 2005-03-13 18:16:24
|
Ahh. Good debugging. It's been a while since I've used the 'null' driver,
I'll duplicate and try and fix it.
On Sun, Mar 13, 2005 at 03:50:34PM +0800, ?????x wrote:
> Hi James,
>=20
> I find a small bug in quagga-mpls/zebra/mpls_null.c.
> Because my kernel has no mpls enhancement. So I modify "mpls_method =3D m=
pls_netlink.o" to "mpls_method =3D mpls_null.o".
> But when I compile, there was an error.
> Finally I find out there was a "missing part" (shown below) in "quagga-mp=
ls/zebra/mpls_null.c". After I add this part. Then I pass the compile.
> int
> mpls_ctrl_tunnel_register(struct interface *ifp, int update)
>=20
> {
>=20
> return 0;
>=20
> }
>=20
> =20
>=20
> int
>=20
> mpls_ctrl_tunnel_unregister(struct interface *ifp)
>=20
> {
>=20
> return 0;
>=20
> }
>=20
>=20
>=20
> Back to the problem (can't run zebra) I am facing now.
> I try to use GDB. Below is some information from GDB.
>=20
> [root@localhost zebra]# gdb zebra
> GNU gdb Red Hat Linux (6.1post-1.20040607.41rh)
> Copyright 2004 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you =
are
> welcome to change it and/or distribute copies of it under certain conditi=
ons.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB. Type "show warranty" for detail=
s.
> This GDB was configured as "i386-redhat-linux-gnu"...Using host libthread=
_db library "/lib/tls/libthread_db.so.1".
>=20
> (gdb)
>=20
> (gdb) run
> Starting program: /root/tim/testing/newtry/quagga-mpls/zebra/zebra
>=20
> Program received signal SIGSEGV, Segmentation fault.
> 0x0037ada3 in strlen () from /lib/tls/libc.so.6
> (gdb) bt
> #0 0x0037ada3 in strlen () from /lib/tls/libc.so.6
> #1 0x0037aae5 in strdup () from /lib/tls/libc.so.6
> #2 0x08066275 in zstrdup (type=3D0, str=3D0x0) at memory.c:115
> #3 0x08058d94 in mpls_out_segment_register (out=3D0xfee52bb4) at mpls_li=
b.c:219
> #4 0x0805c0f3 in netlink_mpls (snl=3D0xfee52c74, h=3D0x1) at rt_netlink.=
c:1844
> #5 0x0805a8f4 in netlink_parse_info (filter=3D0x805c04e <netlink_mpls>,
> nl=3D0x807f548, a=3D0xfee52c84, size=3D16384) at rt_netlink.c:435
> #6 0x0805c24d in mpls_read () at rt_netlink.c:1929
> #7 0x0804cae7 in main (argc=3D1, argv=3D0xfee56de4) at main.c:341
>=20
>=20
> Thanks
>=20
> Timothy
> 2005.03.13
--=20
James R. Leu
jl...@mi...
|
533 messages has been excluded from this view by a project administrator.
