Thread: [mpls-linux-general] Mpls installation problems with the new version.
Status: Beta
Brought to you by:
jleu
From: Oliver H. <ho...@tm...> - 2002-07-31 14:35:58
Attachments:
obiwan.eth1.output
|
Hi, I try to install a mpls network, in a first step with two PCs. I use kernel 2.4.18 and the mpls version 1.152, July 27, 2002. PC1: obiwan (192.168.2.1) <---> PC2: amidala (192.168.2.2) Each PC has a network card with 4 slots (to build up a bigger network later). --- config section start --- Configuration of PC1: Label 21 (push and send) and label 22 (receive and push) [root@obiwan root]# cat /proc/net/mpls_labelspace eth1 0 12 [root@obiwan root]# cat /proc/net/mpls_in 0x40005400 189/21369/0 gen 21 0 1 POP PEEK [root@obiwan root]# cat /proc/net/mpls_out 0x00000004 666/51158/0 2 PUSH(gen 22) SET(eth1,192.168.2.2) [root@obiwan root]# /mnt/image/pkg/ip route add 192.168.2.2/32 via 192.168.2.2 lsp 0x00000004 [root@obiwan root]# netstat -rn Kernel IP Routentabelle Ziel Router Genmask Flags MSS Fenster irtt Iface 192.168.2.2 192.168.2.2 255.255.255.255 UGH 40 0 0 eth1 Configuration of PC2: [root@amidala root]# cat /proc/net/mpls_labelspace eth0 0 11 [root@amidala root]# cat /proc/net/mpls_in 0x40005800 738/59582/0 gen 22 0 1 POP PEEK [root@amidala root]# cat /proc/net/mpls_out 0x00000012 238/25613/0 2 PUSH(gen 21) SET(eth0,192.168.2.1) [root@amidala root]# /mnt/image/pkg/ip route add 192.168.2.1/32 via 192.168.2.1 lsp 0x00000012 [root@amidala root]# netstat -rn Kernel IP Routentabelle Ziel Router Genmask Flags MSS Fenster irtt Iface 192.168.2.1 192.168.2.1 255.255.255.255 UGH 40 0 0 eth0 --- config section stop --- Now I tried to see, that I am really using MPLS. For that reason I have installed ethereal. Now I have the following problems: If I ping obiwan I get the following output of tethereal: 2615.091161 02:01:08:00:36:94 -> c0:a8:02:02:c0:a8 0xb909 Ethernet II 2615.091215 obiwan-2-1 -> amidala ICMP Echo (ping) reply What the hell is 02:01:08:00:36:94? It looks like a MAC address, but we have no cards with these address. The answer of obiwan looks good. I had a look on it with ethereal and it looks like a "normal" mpls header. But the ping request is bad. With ethereal I saw, that there are 6 MPLS-Headers are included. But why? I appended the output of ethereal. The second thing is, that I am able to ping obiwan, but a traceroute to the same IP (only one hop of course) generates no output. I have to press cntl-c. [root@amidala root]# traceroute 192.168.2.1 traceroute to 192.168.2.1 (192.168.2.1), 30 hops max, 38 byte packets [root@amidala root]# So whats wrong? Maybe I have patched the wrong version of iproute2? I hope someone can help me. Thanks. Oliver |
From: James R. L. <jl...@mi...> - 2002-07-31 14:55:50
|
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/mpls-linux/mpls-linux/NOTE?rev=1.5.2.2&only_with_tag=mpls-linux_1_152&content-type=text/vnd.viewcvs-markup On Wed, Jul 31, 2002 at 04:35:47PM +0200, Oliver Horvath wrote: > Hi, > > I try to install a mpls network, in a first step with two PCs. > I use kernel 2.4.18 and the mpls version 1.152, July 27, 2002. > > PC1: obiwan (192.168.2.1) <---> PC2: amidala (192.168.2.2) > > Each PC has a network card with 4 slots (to build up a bigger network > later). > > --- config section start --- > > Configuration of PC1: > Label 21 (push and send) and label 22 (receive and push) > > [root@obiwan root]# cat /proc/net/mpls_labelspace > eth1 0 12 > [root@obiwan root]# cat /proc/net/mpls_in > 0x40005400 189/21369/0 gen 21 0 1 POP PEEK > [root@obiwan root]# cat /proc/net/mpls_out > 0x00000004 666/51158/0 2 PUSH(gen 22) SET(eth1,192.168.2.2) > [root@obiwan root]# /mnt/image/pkg/ip route add 192.168.2.2/32 via > 192.168.2.2 lsp 0x00000004 > [root@obiwan root]# netstat -rn > Kernel IP Routentabelle > Ziel Router Genmask Flags MSS Fenster irtt Iface > 192.168.2.2 192.168.2.2 255.255.255.255 UGH 40 0 0 eth1 > > > Configuration of PC2: > [root@amidala root]# cat /proc/net/mpls_labelspace > eth0 0 11 > [root@amidala root]# cat /proc/net/mpls_in > 0x40005800 738/59582/0 gen 22 0 1 POP PEEK > [root@amidala root]# cat /proc/net/mpls_out > 0x00000012 238/25613/0 2 PUSH(gen 21) SET(eth0,192.168.2.1) > [root@amidala root]# /mnt/image/pkg/ip route add 192.168.2.1/32 via > 192.168.2.1 lsp 0x00000012 > [root@amidala root]# netstat -rn > Kernel IP Routentabelle > Ziel Router Genmask Flags MSS Fenster irtt Iface > 192.168.2.1 192.168.2.1 255.255.255.255 UGH 40 0 0 eth0 > > --- config section stop --- > > Now I tried to see, that I am really using MPLS. For that reason I have > installed ethereal. > > Now I have the following problems: > > If I ping obiwan I get the following output of tethereal: > > 2615.091161 02:01:08:00:36:94 -> c0:a8:02:02:c0:a8 0xb909 Ethernet II > 2615.091215 obiwan-2-1 -> amidala ICMP Echo (ping) reply > > What the hell is 02:01:08:00:36:94? It looks like a MAC address, but we > have no cards with these address. > The answer of obiwan looks good. I had a look on it with ethereal and it > looks like a "normal" mpls header. > But the ping request is bad. With ethereal I saw, that there are 6 > MPLS-Headers are included. But why? I appended the output of ethereal. > > The second thing is, that I am able to ping obiwan, but a traceroute to > the same IP (only one hop of course) generates no output. I have to > press cntl-c. > [root@amidala root]# traceroute 192.168.2.1 > traceroute to 192.168.2.1 (192.168.2.1), 30 hops max, 38 byte packets > > [root@amidala root]# > > So whats wrong? > Maybe I have patched the wrong version of iproute2? > > > I hope someone can help me. > Thanks. > > Oliver > -- James R. Leu |
From: Oliver H. <ho...@tm...> - 2002-07-31 15:07:11
|
Am Mit, 2002-07-31 um 17.53 schrieb James R. Leu: > http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/mpls-linux/mpls-linux/NOTE= ?rev=3D1.5.2.2&only_with_tag=3Dmpls-linux_1_152&content-type=3Dtext/vnd.vie= wcvs-markup ok. That means, that MPLS is working well? Nice! :] But why I am not able to let a traceroute running? Ping works well, but traceroute has not output? (see mailing below) >=20 > On Wed, Jul 31, 2002 at 04:35:47PM +0200, Oliver Horvath wrote: > > Hi, > >=20 > > I try to install a mpls network, in a first step with two PCs. > > I use kernel 2.4.18 and the mpls version 1.152, July 27, 2002. > >=20 > > PC1: obiwan (192.168.2.1) <---> PC2: amidala (192.168.2.2) > >=20 > > Each PC has a network card with 4 slots (to build up a bigger network > > later). > >=20 > > --- config section start --- > >=20 > > Configuration of PC1: > > Label 21 (push and send) and label 22 (receive and push)=20 > >=20 > > [root@obiwan root]# cat /proc/net/mpls_labelspace=20 > > eth1 0 12 > > [root@obiwan root]# cat /proc/net/mpls_in =20 > > 0x40005400 189/21369/0 gen 21 0 1 POP PEEK=20 > > [root@obiwan root]# cat /proc/net/mpls_out=20 > > 0x00000004 666/51158/0 2 PUSH(gen 22) SET(eth1,192.168.2.2)=20 > > [root@obiwan root]# /mnt/image/pkg/ip route add 192.168.2.2/32 via > > 192.168.2.2 lsp 0x00000004 > > [root@obiwan root]# netstat -rn =20 > > Kernel IP Routentabelle > > Ziel Router Genmask Flags MSS Fenster irtt Iface > > 192.168.2.2 192.168.2.2 255.255.255.255 UGH 40 0 0 eth1 > >=20 > >=20 > > Configuration of PC2: > > [root@amidala root]# cat /proc/net/mpls_labelspace=20 > > eth0 0 11 > > [root@amidala root]# cat /proc/net/mpls_in =20 > > 0x40005800 738/59582/0 gen 22 0 1 POP PEEK=20 > > [root@amidala root]# cat /proc/net/mpls_out=20 > > 0x00000012 238/25613/0 2 PUSH(gen 21) SET(eth0,192.168.2.1)=20 > > [root@amidala root]# /mnt/image/pkg/ip route add 192.168.2.1/32 via > > 192.168.2.1 lsp 0x00000012 > > [root@amidala root]# netstat -rn > > Kernel IP Routentabelle > > Ziel Router Genmask Flags MSS Fenster irtt Iface > > 192.168.2.1 192.168.2.1 255.255.255.255 UGH 40 0 0 eth0 > >=20 > > --- config section stop --- > >=20 > > Now I tried to see, that I am really using MPLS. For that reason I have > > installed ethereal. > >=20 > > Now I have the following problems: > >=20 > > If I ping obiwan I get the following output of tethereal: > >=20 > > 2615.091161 02:01:08:00:36:94 -> c0:a8:02:02:c0:a8 0xb909 Ethernet II > > 2615.091215 obiwan-2-1 -> amidala ICMP Echo (ping) reply > >=20 > > What the hell is 02:01:08:00:36:94? It looks like a MAC address, but we > > have no cards with these address. > > The answer of obiwan looks good. I had a look on it with ethereal and i= t > > looks like a "normal" mpls header. > > But the ping request is bad. With ethereal I saw, that there are 6 > > MPLS-Headers are included. But why? I appended the output of ethereal. > >=20 > > The second thing is, that I am able to ping obiwan, but a traceroute to > > the same IP (only one hop of course) generates no output. I have to > > press cntl-c. > > [root@amidala root]# traceroute 192.168.2.1 > > traceroute to 192.168.2.1 (192.168.2.1), 30 hops max, 38 byte packets > >=20 > > [root@amidala root]# > >=20 > > So whats wrong? > > Maybe I have patched the wrong version of iproute2? > >=20 > >=20 > > I hope someone can help me. > > Thanks. > >=20 > > Oliver > >=20 >=20 >=20 >=20 > --=20 > James R. Leu >=20 >=20 > ------------------------------------------------------- > This sf.net email is sponsored by: Dice - The leading online job board > for high-tech professionals. Search and apply for tech jobs today! > http://seeker.dice.com/seeker.epl?rel_code=3D31 > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general >=20 --=20 Dipl.-Inf. Oliver Horvath Johann Wolfgang Goethe-Universit=E4t Tel: 069/798-28358 (Fax: 23340) Institut f=FCr Informatik - Telematik |
From: James R. L. <jl...@mi...> - 2002-07-31 15:24:06
|
On Wed, Jul 31, 2002 at 05:07:04PM +0200, Oliver Horvath wrote: > Am Mit, 2002-07-31 um 17.53 schrieb James R. Leu: > > http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/mpls-linux/mpls-linux/= NOTE?rev=3D1.5.2.2&only_with_tag=3Dmpls-linux_1_152&content-type=3Dtext/v= nd.viewcvs-markup > ok. That means, that MPLS is working well? Nice! :] > But why I am not able to let a traceroute running? > Ping works well, but traceroute has not output? (see mailing below) Handling of packets with TTL <=3D 1 is broken right now. The result is t= hat the packet gets dropped. I think I know how I'm going to handle it: Check if this is the botton of the stack, if not drop the packet. Check the protocol associated with the ILM, if it is not IP, drop it. If it is IP, generate an ICMP message, and forward it along the LSP, the egress can deal with the ICMP message. There is probably a special case where it is the egress of the LSP that finds a TTL <=3D 1 that needs to be handled differntly. Besides that, does this sound like the right behavior? Jim >=20 > >=20 > > On Wed, Jul 31, 2002 at 04:35:47PM +0200, Oliver Horvath wrote: > > > Hi, > > >=20 > > > I try to install a mpls network, in a first step with two PCs. > > > I use kernel 2.4.18 and the mpls version 1.152, July 27, 2002. > > >=20 > > > PC1: obiwan (192.168.2.1) <---> PC2: amidala (192.168.2.2) > > >=20 > > > Each PC has a network card with 4 slots (to build up a bigger netwo= rk > > > later). > > >=20 > > > --- config section start --- > > >=20 > > > Configuration of PC1: > > > Label 21 (push and send) and label 22 (receive and push)=20 > > >=20 > > > [root@obiwan root]# cat /proc/net/mpls_labelspace=20 > > > eth1 0 12 > > > [root@obiwan root]# cat /proc/net/mpls_in =20 > > > 0x40005400 189/21369/0 gen 21 0 1 POP PEEK=20 > > > [root@obiwan root]# cat /proc/net/mpls_out=20 > > > 0x00000004 666/51158/0 2 PUSH(gen 22) SET(eth1,192.168.2.2)=20 > > > [root@obiwan root]# /mnt/image/pkg/ip route add 192.168.2.2/32 via > > > 192.168.2.2 lsp 0x00000004 > > > [root@obiwan root]# netstat -rn =20 > > > Kernel IP Routentabelle > > > Ziel Router Genmask Flags MSS Fenster irtt Ifa= ce > > > 192.168.2.2 192.168.2.2 255.255.255.255 UGH 40 0 0 eth1 > > >=20 > > >=20 > > > Configuration of PC2: > > > [root@amidala root]# cat /proc/net/mpls_labelspace=20 > > > eth0 0 11 > > > [root@amidala root]# cat /proc/net/mpls_in =20 > > > 0x40005800 738/59582/0 gen 22 0 1 POP PEEK=20 > > > [root@amidala root]# cat /proc/net/mpls_out=20 > > > 0x00000012 238/25613/0 2 PUSH(gen 21) SET(eth0,192.168.2.1)=20 > > > [root@amidala root]# /mnt/image/pkg/ip route add 192.168.2.1/32 via > > > 192.168.2.1 lsp 0x00000012 > > > [root@amidala root]# netstat -rn > > > Kernel IP Routentabelle > > > Ziel Router Genmask Flags MSS Fenster irtt I= face > > > 192.168.2.1 192.168.2.1 255.255.255.255 UGH 40 0 0 e= th0 > > >=20 > > > --- config section stop --- > > >=20 > > > Now I tried to see, that I am really using MPLS. For that reason I = have > > > installed ethereal. > > >=20 > > > Now I have the following problems: > > >=20 > > > If I ping obiwan I get the following output of tethereal: > > >=20 > > > 2615.091161 02:01:08:00:36:94 -> c0:a8:02:02:c0:a8 0xb909 Ethernet = II > > > 2615.091215 obiwan-2-1 -> amidala ICMP Echo (ping) reply > > >=20 > > > What the hell is 02:01:08:00:36:94? It looks like a MAC address, bu= t we > > > have no cards with these address. > > > The answer of obiwan looks good. I had a look on it with ethereal a= nd it > > > looks like a "normal" mpls header. > > > But the ping request is bad. With ethereal I saw, that there are 6 > > > MPLS-Headers are included. But why? I appended the output of ethere= al. > > >=20 > > > The second thing is, that I am able to ping obiwan, but a tracerout= e to > > > the same IP (only one hop of course) generates no output. I have to > > > press cntl-c. > > > [root@amidala root]# traceroute 192.168.2.1 > > > traceroute to 192.168.2.1 (192.168.2.1), 30 hops max, 38 byte packe= ts > > >=20 > > > [root@amidala root]# > > >=20 > > > So whats wrong? > > > Maybe I have patched the wrong version of iproute2? > > >=20 > > >=20 > > > I hope someone can help me. > > > Thanks. > > >=20 > > > Oliver > > >=20 > >=20 > >=20 > >=20 > > --=20 > > James R. Leu > >=20 > >=20 > > ------------------------------------------------------- > > This sf.net email is sponsored by: Dice - The leading online job boar= d > > for high-tech professionals. Search and apply for tech jobs today! > > http://seeker.dice.com/seeker.epl?rel_code=3D31 > > _______________________________________________ > > mpls-linux-general mailing list > > mpl...@li... > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > >=20 > --=20 > Dipl.-Inf. Oliver Horvath Johann Wolfgang Goethe-Universit=E4t > Tel: 069/798-28358 (Fax: 23340) Institut f=FCr Informatik - Telematik >=20 >=20 >=20 > ------------------------------------------------------- > This sf.net email is sponsored by: Dice - The leading online job board > for high-tech professionals. Search and apply for tech jobs today! > http://seeker.dice.com/seeker.epl?rel_code1 > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu |
From: Oliver H. <ho...@tm...> - 2002-07-31 16:22:05
|
Am Mit, 2002-07-31 um 18.21 schrieb James R. Leu: > On Wed, Jul 31, 2002 at 05:07:04PM +0200, Oliver Horvath wrote: > > Am Mit, 2002-07-31 um 17.53 schrieb James R. Leu: > > > http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/mpls-linux/mpls-linux/NOTE?rev=1.5.2.2&only_with_tag=mpls-linux_1_152&content-type=text/vnd.viewcvs-markup > > ok. That means, that MPLS is working well? Nice! :] > > But why I am not able to let a traceroute running? > > Ping works well, but traceroute has not output? (see mailing below) > > Handling of packets with TTL <= 1 is broken right now. The result is that the > packet gets dropped. I think I know how I'm going to handle it: > > Check if this is the botton of the stack, if not drop the packet. > Check the protocol associated with the ILM, if it is not IP, drop it. > If it is IP, generate an ICMP message, and forward it along the LSP, the > egress can deal with the ICMP message. > > There is probably a special case where it is the egress of the LSP that > finds a TTL <= 1 that needs to be handled differntly. Besides that, > does this sound like the right behavior? Sounds good for me. I tested ethereal in the middle of a lsp (no edge router) and there it works well. Do you have a idea, if there is a chance to monitor the edge routers correctly too? I defined a wrong outgoing lable on a edge router, and I saw two things: a) Ping is still working, but why? If mpls didn't know the incoming label, it pops the label off and tries to find a working label in the outgoing labels? Or is the label poped off and forwarded by normal ip? b) Tracerout stops at the middle of the lsp (3 PCs). Why it is not forwarded like the pings? Oliver |
From: James R. L. <jl...@mi...> - 2002-07-31 16:53:12
|
> > Handling of packets with TTL <= 1 is broken right now. The result is that the > > packet gets dropped. I think I know how I'm going to handle it: > > > > Check if this is the botton of the stack, if not drop the packet. > > Check the protocol associated with the ILM, if it is not IP, drop it. > > If it is IP, generate an ICMP message, and forward it along the LSP, the > > egress can deal with the ICMP message. > > > > There is probably a special case where it is the egress of the LSP that > > finds a TTL <= 1 that needs to be handled differntly. Besides that, > > does this sound like the right behavior? > > Sounds good for me. > > I tested ethereal in the middle of a lsp (no edge router) and there it > works well. > Do you have a idea, if there is a chance to monitor the edge routers > correctly too? For this to work correctly I would have to make a complete copy of the packet and modify the copy, not the original. This would be slow, so I don't want to do it as part of the normal path. If people would really like to see "correct" MPLS packets at the LERs maybe I can figure out a way to tell if someone else has a handle to the packet and change the path accordingly. If you would like a detailed description of why it workes the way it does I can walk you through the code. > I defined a wrong outgoing lable on a edge router, and I saw two things: > a) Ping is still working, but why? If mpls didn't know the incoming > label, it pops the label off and tries to find a working label in the > outgoing labels? Or is the label poped off and forwarded by normal ip? I haven't done any negative testing recently, I will try your test tonight. It shouldn't have worked. > b) Tracerout stops at the middle of the lsp (3 PCs). Why it is not > forwarded like the pings? *shrug* Let me figure out what the ping _worked_ first :-) Jim -- James R. Leu |