Re: [mpls-linux-general] LDP + ZEBRA HELP ! config inside ...
Status: Beta
Brought to you by:
jleu
|
From: James R. L. <jl...@mi...> - 2003-03-19 15:53:39
|
The first thing I would do is remove all of your router-id statments (from zebra and ospf). Let the zebra router-id mechanism pick it. On Wed, Mar 19, 2003 at 03:22:54PM -0000, Nicolas Tempestini wrote: > Hello, > > I now try to set up a very small network between 2 machines and try to use zebra, mplsd and LDP. > > Here is my config : > > > |--------------| |--------------| > -------- | comp 1 |--------------------| comp2 |-------- > |--------------| |--------------| > > > comp1 is running eth0 with adr Ip xxx.xxx.214.131 > comp2 is running eth0 with adr Ip xxx.xxx.214.132 > > All other interfaces are down > > comp1 is running dummy0 interface on 192.168.0.2/32 > comp2 is running dummy0 interface on 192.168.0.3/32 > > Zebra and opsfd are running perfectly. > > xxx.xxx.214.129 is a router running Ospf. > > here are my zebra daemons files : > > > zebra.conf > -------------- > ! Zebra configuration saved from vty > ! 2003/03/19 12:56:14 > > ! > > hostname Router > > password zebra > > enable password zebra > > log file /usr/local/etc/zebra.log > > ! > > interface lo > > ! > > interface eth0 > > ! > > interface eth1 > > shutdown > > ! > > interface dummy0 > > ip address 192.168.0.2/32 > > ! > > router-id xxx.xxx.214.131 > > ! > > line vty > > ! > > ospfd.conf > -------------- > > ! Zebra configuration saved from vty > > ! 2003/03/14 14:27:47 > > ! > > hostname ospfd > > password zebra > > enable password zebra > > log stdout > > ! > > interface lo > > ! > > interface dummy0 > > ! > > interface eth0 > > ! > > interface eth1 > > ! > > router ospf > > passive-interface dummy0 > > ospf router-id xxx.xxx.214.131 > > ospf abr-type cisco > > network xxx.xxx.214.128/26 area 1 > > network 192.168.0.0/24 area 1 > > ! > > line vty > > ! > > mplsd.conf > > ! Zebra configuration saved from vty > > ! 2003/03/17 11:07:20 > > ! > > hostname uml-1 > > password mpls > > enable password mpls > > ! > > mpls ldp > > egress connected > > ! > > interface lo > > no mpls labelspace > > ! > > interface eth0 > > mpls labelspace 1 > > mpls ldp > > ! > > interface eth1 > > mpls labelspace 2 > > mpls ldp > > ! > > interface dummy0 > > ! > > line vty > > Here are the output of some mplsd commands : > > show ldp neighbor > > Peer LDP Ident: xxx.xxx.214.132:0; Local LDP Ident: xxx.xxx.214.131:1 > > TCP connection: n/a > > State: discovery; Msgs sent/recv: -/-; > > Up time: - > > LDP discovery sources: > > eth0 > > Peer LDP Ident: xxx.xxx.214.132:0; Local LDP Ident: xxx.xxx.214.131:1 > > TCP connection: n/a > > State: discovery; Msgs sent/recv: -/-; > > Up time: - > > LDP discovery sources: > > eth0 > > Peer LDP Ident: xxx.xxx.214.132:0; Local LDP Ident: xxx.xxx.214.131:1 > > TCP connection: n/a > > State: discovery; Msgs sent/recv: -/-; > > Up time: - > > LDP discovery sources: > > eth0 > > > > uml-1# show ldp database > > no labeling info has been exchanged > > uml-1# show lsr-id > > lsr-id: xxx.xxx.214.131 > > > > I try to set up static mpls in-segment and out-segment between those machines using mplsd , here are /proc/net/mpls_*: > > mpls_in > > :::::::::::::: > > 0x008ae005 0/0/0 gen 2222 1 1 POP PEEK > > :::::::::::::: > > mpls_labelspace > > :::::::::::::: > > eth0 1 37 > > eth1 2 3 > > dummy0 0 3 > > :::::::::::::: > > mpls_out > > :::::::::::::: > > 0x00000010 0/0/0 1 PUSH(gen 1234) SET(eth0,137.194.214.132) > > :::::::::::::: > > mpls_tunnel > > :::::::::::::: > > :::::::::::::: > > mpls_version > > :::::::::::::: > > 01010702 > > > > when i do a 'trace all' in mplsd i see that both lsr are speaking together but there are errors like: > > chpTlv Flags: target = 0 > > OUT: request = 0 > > OUT: res = 0 > > OUT: Hello msg does not have TrAdr TLV > > OUT: csnTlv: > > OUT: Tlv: > > OUT: BaseTlv: uBit = 0 > > OUT: fBit = 0 > > OUT: type = 402 > > OUT: length = 4 > > OUT: csnTlv data: value = 6 > > OUT: HELLO MSG ***END***: > > OUT: Mesg size: 24 (34) > > EXIT: ldp_decode_one_mesg > > ENTER: ldp_state_machine > > OUT: FSM: state 0, event 0 > > ENTER: ldp_state_new_adjacency > > ENTER: ldp_hello_process > > OUT: Hello Recv: entity(1) > > OUT: ldp_hello_process: PASSIVE(27) > > ENTER: ldp_adj_backoff_stop > > EXIT: ldp_adj_backoff_stop > > EXIT: ldp_hello_process > > ENTER: ldp_adj_startup > > EXIT: ldp_adj_startup > > EXIT: ldp_state_new_adjacency > > EXIT: ldp_state_machine > > EXIT: ldp_buf_process > > ENTER: ldp_buf_process > > OUT: ldp_event: LDP_EVENT_UDP_DATA errno = 11 > > EXIT: ldp_buf_process > > EXIT: ldp_event > > > > Can you tell me why There is no label distribution ? and what to do in order to make it work .... ???? > > > > THX a lot ! > > > > > > > > > > > -- James R. Leu |