mpls-linux-general Mailing List for MPLS for Linux (Page 128)
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: Alexandre C. <ale...@re...> - 2002-10-21 22:43:47
|
Here is my mplsd file can anyone tell me whats wrong. Regards. Alex |
|
From: Alexandre C. <ale...@re...> - 2002-10-21 22:39:17
|
I=B4ve just installed ldp-portable 0.250... Before i start the mpls deamon the network is working fine. But when i st= art the mplsd and try to telnet the localhost, it freeze, and i can=B4t type = the password to enter to see what is going on. Does any one has that problem? Any help help would be helpfull. Regards Alex |
|
From: James R. L. <jl...@mi...> - 2002-10-21 16:33:46
|
On Mon, Oct 21, 2002 at 07:05:50PM +0800, Folke Aeon wrote: > hi, J. >=20 > thanks for your answer. but i don;t quite > understand the meaning of "Reference count" > would you please tell me how can i find > more detailed information about it? >=20 > thanks for your precious time :) Pointers to MPLS structure are held by other kernel entities that may be deleted at differnt times then the MPLS structure itself. A common wa= y to handle this is to reference count the structure who's pointer is being held by other entities. So when a kernel entity needs the pointer to the MPLS structure, the refe= rence count is incremented. When the kernel entity no longer need the pointer = to the MPLS structure the reference count is decremented. When the reference co= unt reaches 0, the MPLS structure is deleted. The only reason the reference counters are displayed in the /proc/ output is to help debug cases where MPLS structures are not being cleaned up. >=20 > folke. >=20 >=20 >=20 >=20 > >From: "James R. Leu" <jl...@mi...> > >Reply-To: jl...@mi... > >To: Folke Aeon <aeo...@ho...> > >CC: Subject: Re: [mpls-linux-general] i want to make sure some paramet= ers=20 > >in /proc/net/mpls* > >Date: Sun, 20 Oct 2002 18:29:28 -0500 > > > >On Sat, Oct 19, 2002 at 03:50:30PM +0800, Folke Aeon wrote: > > > hi, all > > > > > > will anyone tell me the meaning of the following parameters in > > > /proc/net/mpls* :? > > > > > > [root@localhost net]# cat mpls_in > > > 0x40004000 4/256/0 gen 16 0 1 POP PEEK > > > + + + > > > | | | > > > | | +------what does this > > > | | mean then? > > > | | > > > | label space? > > > | > > > what does this zero mean? packets > > > dropped? or bytest dropped? > > > >Dropped packets. > > > >The '1' is the reference count. >=20 >=20 > _________________________________________________________________ > Get faster connections=A0-- switch to=A0MSN Internet Access!=20 > http://resourcecenter.msn.com/access/plans/default.asp >=20 >=20 >=20 > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu |
|
From: James R. L. <jl...@mi...> - 2002-10-21 16:26:52
|
On Mon, Oct 21, 2002 at 05:07:42PM +0530, Amritpal Singh wrote: > Hi James, > > Ya the message "mpls_bind_out2fec: couldn't find the FIB_NODE(0)" is gone now > after having added the host-specific route. > But still the problem of counters in /proc/net/mpls_* being zero persists. > > I found out that having configured a LSP between B and C only, the counters increment only when I ping > from A to C or from B to C or D. This is strange because I was testing MPLS by pinging D from A. > > > A---------------B-----------C---------------------D > > Can you enlighten me on this ? Why is ping from A to D not taking the MPLS path ? Are you creating cross connects on B and C? mplsadm2 -B -I gen:16:0 -O 0x4 > > Regards > Amrit > ------------------------------ > ----- Original Message ----- > From: James R. Leu > To: Amritpal Singh > Cc: mpl...@li... > Sent: Friday, October 18, 2002 11:14 PM > Subject: Re: [mpls-linux-general] SOS.....MPLS forwarding not working...URGENT > > > The first change I suggest you make is to switch to the latest version > of mpls-linux-1.1. Many things have changed since 1.0, and I do not have > the time to downgrade to mpls-linux-1.0 to test/verify any of the > mpls-linux-1.0 command you are try, but...... > > ... since it seems like you need to stay with mpls-linux-1.0 the only thing > I can see wrong (from a quick skim of your output) is that when specifing > -f <prefix>, the prefix must have an EXACT match in your routing table. > > In your example below, 10.2.0.2/32 must show up in your routing table. > The "mpls_bind_out2fec: couldn't find the FIB_NODE(0)" message in > the 'dmesg' output is what tells me that 10.2.0.2/32 is not in your > routing table. > > Jim > > On Fri, Oct 18, 2002 at 06:20:46PM +0530, Amritpal Singh wrote: > > > > Hi All, > > > > The following is the setup in which I am trying to test MPLS. > > > > > > -------- ----------------- ----------------- --------- > > | |------------------ | |------------------| |------------------- | | > > | A | | B | | C | | D | > > -------- ----------------- ----------------- --------- > > 10.1.1.2 10.1.1.1 10.2.0.1 10.2.0.2 10.2.1.1 10.2.1.10 > > eth0 eth0 eth1 eth0 eth1 eth0 > > > > I am running mpls-linux-1.0 on a Red Hat Linux 7.2 > > All the nodes hav the patch applied to them. > > > > Nodes B and C are the ones which should enable me to see label switching. > > So I configure the incoming and outgoing labels on B and C as follows. > > > > **************FOR NODE B ****************************************************** > > [root@ferrari utils]# ./mplsadm -v -A -O gen:16:eth1:ipv4:10.2.0.2 > > > > Out label input: gen:16:eth1:ipv4:10.2.0.2 > > > > ipv4 > > > > [root@ferrari utils]# > > > > [root@ferrari utils]# ./mplsadm -B -O gen:16:eth1 -f 10.2.0.2/32 > > > > [root@ferrari utils]# > > > > [root@ferrari utils]# ./mplsadm -L eth1:0 > > > > [root@ferrari utils]# ./mplsadm -v -A -I gen:17:0 > > > > In label input: gen:17:0 > > > > [root@ferrari utils]# ./mplsadm -d > > > > Debug: Success > > > > [root@ferrari utils]# ./mplsadm -v -B -O gen:16:eth1 -f 10.2.0.2/32 > > > > Out label input: gen:16:eth1 > > > > FEC input: 10.2.0.2/32 > > > > ****************END FOR NODE B *********************************************** > > > > ****************FOR NODE C ***************************************************** > > [root@renault utils]# ./mplsadm -v -A -O gen:17:eth0:ipv4:10.2.0.1 > > > > Out label input: gen:17:eth0:ipv4:10.2.0.1 > > > > ipv4 > > > > [root@renault utils]# > > > > [root@renault utils]# ./mplsadm -v -B -O gen:17:eth0 -f 10.2.0.1/32 > > > > Out label input: gen:17:eth0 > > > > FEC input: 10.2.0.1/32 > > > > [root@renault utils]# ./mplsadm -L eth0:0 > > > > [root@renault utils]# ./mplsadm -v -A -I gen:16:0 > > > > In label input: gen:16:0 > > > > [root@renault utils]# ./mplsadm -L eth1:0 > > > > [root@renault utils]# > > > > *****************************END FOR NODE C ************************************* > > > > > > > > I see all the valid entries in all the mpls_* files in /proc/net as shown below. > > > > **********FOR NODE B ************************************************************** > > [root@ferrari Desktop]# more /proc/net/mpls_* > > > > :::::::::::::: > > > > /proc/net/mpls_fec > > > > :::::::::::::: > > > > :::::::::::::: > > > > /proc/net/mpls_in > > > > :::::::::::::: > > > > 40004400 0/0/0 gen 17 0 POP PEEK > > > > :::::::::::::: > > > > /proc/net/mpls_labelspace > > > > :::::::::::::: > > > > lo 0 > > > > eth1 0 > > > > :::::::::::::: > > > > /proc/net/mpls_out > > > > :::::::::::::: > > > > 40004003 0/0/0 PUSH(gen 16) SET(eth1,10.2.0.2) > > > > :::::::::::::: > > > > /proc/net/mpls_tunnel > > > > :::::::::::::: > > > > *************END FOR NODE B *************************************************** > > ***************FOR NODE C ****************************************************** > > [root@renault root]# more /proc/net/mpls_* > > > > :::::::::::::: > > > > /proc/net/mpls_fec > > > > :::::::::::::: > > > > :::::::::::::: > > > > /proc/net/mpls_in > > > > :::::::::::::: > > > > 40004000 0/0/0 gen 16 0 POP PEEK > > > > :::::::::::::: > > > > /proc/net/mpls_labelspace > > > > :::::::::::::: > > > > lo 0 > > > > eth0 0 > > > > :::::::::::::: > > > > /proc/net/mpls_out > > > > :::::::::::::: > > > > 40004403 0/0/0 PUSH(gen 17) SET(eth0,10.2.0.1) > > > > :::::::::::::: > > > > /proc/net/mpls_tunnel > > > > :::::::::::::: > > > > > > > > But when I try to ping D from A I donot see the label switching happening at all.Ping works fine. > > > > But the counters in mpls_in and mpls_out remain 0.Plus I donot see anything in mpls_fec.Should I ? > > > > I run ethereal on B and C to see incoming/outgoing labels on the packet but I find that > > there is absolutely no label on the packets. > > > > The last few lines of output of "DMESG" are shown below . > > ************************FOR NODE B ***************************************** > > MPLS version 0.996 11/21/2001 jl...@mi... > > > > MPLS Tunnel interface > > > > ds: no socket drivers loaded! > > > > VFS: Mounted root (ext2 filesystem) readonly. > > > > Freeing unused kernel memory: 224k freed > > > > Adding Swap: 153176k swap-space (priority -1) > > > > mpls_prep_out2fec: enter > > > > Label GEN 16 > > > > Key GEN 16 3 > > > > mpls_fill_key_res: enter > > > > mpls_fill_key_res: exit > > > > mpls_bind_out2fec: couldn't find the FIB_NODE(0) > > > > mpls_bind_out2fec: exit > > > > mpls_set_labelspace: enter > > > > mpls_set_labelspace: labelspace(0) > > > > mpls_set_labelspace: exit > > > > *****************************END FOR NODE B ******************************* > > ***************************** FOR NODE C ************************************ > > MPLS version 0.996 11/21/2001 jl...@mi... > > > > MPLS Tunnel interface > > > > ds: no socket drivers loaded! > > > > VFS: Mounted root (ext2 filesystem) readonly. > > > > Freeing unused kernel memory: 224k freed > > > > Adding Swap: 153176k swap-space (priority -1) > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > *******************************************END FOR NODE C ******************************** > > > > I cannot upgrade to the latest version available because of time constraints.I had downloaded this > > version some months back and I have to show MPLS working with this version only. > > > > I will be very thankful to all the people who respond and help me in any way. > > > > I am looking forward to your comments on this. > > > > P.S: Last query: Do I need to "decode" the ICMP packet received at node C by ethereal into a MPLS packet. > > As I see the raw packet received at the ethereal there is no MPLS header. But there is an option in ethereal > > to decode the packet in a number of way. So should I decode it at the link layer as MPLS packet to see the header? > > > > Thanks again > > Amrit > > > > -------------------------------- > > Amritpal Singh > > Infosys Technologies Ltd. > > Bangalore - 561229 > > Tel No : 4166308 > > --------------------------------- > > > -- > James R. Leu > -- James R. Leu |
|
From: Folke A. <aeo...@ho...> - 2002-10-21 11:07:31
|
hi, J. thanks for your answer. but i don;t quite understand the meaning of "Reference count" would you please tell me how can i find more detailed information about it? thanks for your precious time :) folke. >From: "James R. Leu" <jl...@mi...> >Reply-To: jl...@mi... >To: Folke Aeon <aeo...@ho...> >CC: Subject: Re: [mpls-linux-general] i want to make sure some parameters >in /proc/net/mpls* >Date: Sun, 20 Oct 2002 18:29:28 -0500 > >On Sat, Oct 19, 2002 at 03:50:30PM +0800, Folke Aeon wrote: > > hi, all > > > > will anyone tell me the meaning of the following parameters in > > /proc/net/mpls* :? > > > > [root@localhost net]# cat mpls_in > > 0x40004000 4/256/0 gen 16 0 1 POP PEEK > > + + + > > | | | > > | | +------what does this > > | | mean then? > > | | > > | label space? > > | > > what does this zero mean? packets > > dropped? or bytest dropped? > >Dropped packets. > >The '1' is the reference count. _________________________________________________________________ Get faster connections -- switch to MSN Internet Access! http://resourcecenter.msn.com/access/plans/default.asp |
|
From: James R. L. <jl...@mi...> - 2002-10-20 22:28:09
|
On Sat, Oct 19, 2002 at 03:50:30PM +0800, Folke Aeon wrote: > hi, all > > will anyone tell me the meaning of the following parameters in > /proc/net/mpls* :? > > [root@localhost net]# cat mpls_in > 0x40004000 4/256/0 gen 16 0 1 POP PEEK > + + + > | | | > | | +------what does this > | | mean then? > | | > | label space? > | > what does this zero mean? packets > dropped? or bytest dropped? Dropped packets. The '1' is the reference count. > > > [root@localhost net]# cat mpls_out > 0x00000002 7837/477621/0 3 PUSH(gen 230) SET(eth1,11.0.2.2) > + > | > | > what does this mean? Reference count. > > [root@localhost net]# cat mpls_labelspace > eth0 0 11 > + > | > | > and what about this? Reference count. > > thanks in advacne > > folke. > > _________________________________________________________________ > Internet access plans that fit your lifestyle -- join MSN. > http://resourcecenter.msn.com/access/plans/default.asp > > > > ------------------------------------------------------- > This sf.net email is sponsored by: > Access Your PC Securely with GoToMyPC. Try Free Now > https://www.gotomypc.com/s/OSND/DD > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu |
|
From: <no...@me...> - 2002-10-19 10:29:23
|
> Send mpls-linux-general mailing list submissions to > mpl...@li... > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > or, via email, send a message with subject or body 'help' to > mpl...@li... > > You can reach the person managing the list at > mpl...@li... > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of mpls-linux-general digest..." > > > Today's Topics: > > 1. Re: SOS.....MPLS forwarding not working...URGENT (James R. Leu) > 2. Re: make problem with mplsadm on fresh Mdk9.0 distro (James R. Leu) > 3. Re: Re: Dynamic LSP (James R. Leu) > 4. Re: LDP config... confused! (James R. Leu) > 5. Re: ldp with ip unnumbered (James R. Leu) > > --__--__-- > > Message: 1 > Date: Fri, 18 Oct 2002 12:44:41 -0500 > From: "James R. Leu" <jl...@mi...> > To: Amritpal Singh <amr...@in...> > Cc: mpl...@li... > Subject: Re: [mpls-linux-general] SOS.....MPLS forwarding not working...URGENT > Reply-To: jl...@mi... > Organization: none > > The first change I suggest you make is to switch to the latest version > of mpls-linux-1.1. Many things have changed since 1.0, and I do not have > the time to downgrade to mpls-linux-1.0 to test/verify any of the > mpls-linux-1.0 command you are try, but...... > > .... since it seems like you need to stay with mpls-linux-1.0 the only thing > I can see wrong (from a quick skim of your output) is that when specifing > -f <prefix>, the prefix must have an EXACT match in your routing table. > > In your example below, 10.2.0.2/32 must show up in your routing table. > The "mpls_bind_out2fec: couldn't find the FIB_NODE(0)" message in > the 'dmesg' output is what tells me that 10.2.0.2/32 is not in your > routing table. > > Jim > > On Fri, Oct 18, 2002 at 06:20:46PM +0530, Amritpal Singh wrote: > > > > Hi All, > > > > The following is the setup in which I am trying to test MPLS. > > > > > > -------- ----------------- ----------------- --------- > > | |------------------ | |------------------| |------------------- | | > > | A | | B | | C | | D | > > -------- ----------------- ----------------- --------- > > 10.1.1.2 10.1.1.1 10.2.0.1 10.2.0.2 10.2.1.1 10.2.1.10 > > eth0 eth0 eth1 eth0 eth1 eth0 > > > > I am running mpls-linux-1.0 on a Red Hat Linux 7.2 > > All the nodes hav the patch applied to them. > > > > Nodes B and C are the ones which should enable me to see label switching. > > So I configure the incoming and outgoing labels on B and C as follows. > > > > **************FOR NODE B ****************************************************** > > [root@ferrari utils]# ./mplsadm -v -A -O gen:16:eth1:ipv4:10.2.0.2 > > > > Out label input: gen:16:eth1:ipv4:10.2.0.2 > > > > ipv4 > > > > [root@ferrari utils]# > > > > [root@ferrari utils]# ./mplsadm -B -O gen:16:eth1 -f 10.2.0.2/32 > > > > [root@ferrari utils]# > > > > [root@ferrari utils]# ./mplsadm -L eth1:0 > > > > [root@ferrari utils]# ./mplsadm -v -A -I gen:17:0 > > > > In label input: gen:17:0 > > > > [root@ferrari utils]# ./mplsadm -d > > > > Debug: Success > > > > [root@ferrari utils]# ./mplsadm -v -B -O gen:16:eth1 -f 10.2.0.2/32 > > > > Out label input: gen:16:eth1 > > > > FEC input: 10.2.0.2/32 > > > > ****************END FOR NODE B *********************************************** > > > > ****************FOR NODE C ***************************************************** > > [root@renault utils]# ./mplsadm -v -A -O gen:17:eth0:ipv4:10.2.0.1 > > > > Out label input: gen:17:eth0:ipv4:10.2.0.1 > > > > ipv4 > > > > [root@renault utils]# > > > > [root@renault utils]# ./mplsadm -v -B -O gen:17:eth0 -f 10.2.0.1/32 > > > > Out label input: gen:17:eth0 > > > > FEC input: 10.2.0.1/32 > > > > [root@renault utils]# ./mplsadm -L eth0:0 > > > > [root@renault utils]# ./mplsadm -v -A -I gen:16:0 > > > > In label input: gen:16:0 > > > > [root@renault utils]# ./mplsadm -L eth1:0 > > > > [root@renault utils]# > > > > *****************************END FOR NODE C ************************************* > > > > > > > > I see all the valid entries in all the mpls_* files in /proc/net as shown below. > > > > **********FOR NODE B ************************************************************** > > [root@ferrari Desktop]# more /proc/net/mpls_* > > > > :::::::::::::: > > > > /proc/net/mpls_fec > > > > :::::::::::::: > > > > :::::::::::::: > > > > /proc/net/mpls_in > > > > :::::::::::::: > > > > 40004400 0/0/0 gen 17 0 POP PEEK > > > > :::::::::::::: > > > > /proc/net/mpls_labelspace > > > > :::::::::::::: > > > > lo 0 > > > > eth1 0 > > > > :::::::::::::: > > > > /proc/net/mpls_out > > > > :::::::::::::: > > > > 40004003 0/0/0 PUSH(gen 16) SET(eth1,10.2.0.2) > > > > :::::::::::::: > > > > /proc/net/mpls_tunnel > > > > :::::::::::::: > > > > *************END FOR NODE B *************************************************** > > ***************FOR NODE C ****************************************************** > > [root@renault root]# more /proc/net/mpls_* > > > > :::::::::::::: > > > > /proc/net/mpls_fec > > > > :::::::::::::: > > > > :::::::::::::: > > > > /proc/net/mpls_in > > > > :::::::::::::: > > > > 40004000 0/0/0 gen 16 0 POP PEEK > > > > :::::::::::::: > > > > /proc/net/mpls_labelspace > > > > :::::::::::::: > > > > lo 0 > > > > eth0 0 > > > > :::::::::::::: > > > > /proc/net/mpls_out > > > > :::::::::::::: > > > > 40004403 0/0/0 PUSH(gen 17) SET(eth0,10.2.0.1) > > > > :::::::::::::: > > > > /proc/net/mpls_tunnel > > > > :::::::::::::: > > > > > > > > But when I try to ping D from A I donot see the label switching happening at all.Ping works fine. > > > > But the counters in mpls_in and mpls_out remain 0.Plus I donot see anything in mpls_fec.Should I ? > > > > I run ethereal on B and C to see incoming/outgoing labels on the packet but I find that > > there is absolutely no label on the packets. > > > > The last few lines of output of "DMESG" are shown below . > > ************************FOR NODE B ***************************************** > > MPLS version 0.996 11/21/2001 jl...@mi... > > > > MPLS Tunnel interface > > > > ds: no socket drivers loaded! > > > > VFS: Mounted root (ext2 filesystem) readonly. > > > > Freeing unused kernel memory: 224k freed > > > > Adding Swap: 153176k swap-space (priority -1) > > > > mpls_prep_out2fec: enter > > > > Label GEN 16 > > > > Key GEN 16 3 > > > > mpls_fill_key_res: enter > > > > mpls_fill_key_res: exit > > > > mpls_bind_out2fec: couldn't find the FIB_NODE(0) > > > > mpls_bind_out2fec: exit > > > > mpls_set_labelspace: enter > > > > mpls_set_labelspace: labelspace(0) > > > > mpls_set_labelspace: exit > > > > *****************************END FOR NODE B ******************************* > > ***************************** FOR NODE C ************************************ > > MPLS version 0.996 11/21/2001 jl...@mi... > > > > MPLS Tunnel interface > > > > ds: no socket drivers loaded! > > > > VFS: Mounted root (ext2 filesystem) readonly. > > > > Freeing unused kernel memory: 224k freed > > > > Adding Swap: 153176k swap-space (priority -1) > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > device eth0 entered promiscuous mode > > > > device eth0 left promiscuous mode > > > > *******************************************END FOR NODE C ******************************** > > > > I cannot upgrade to the latest version available because of time constraints.I had downloaded this > > version some months back and I have to show MPLS working with this version only. > > > > I will be very thankful to all the people who respond and help me in any way. > > > > I am looking forward to your comments on this. > > > > P.S: Last query: Do I need to "decode" the ICMP packet received at node C by ethereal into a MPLS packet. > > As I see the raw packet received at the ethereal there is no MPLS header. But there is an option in ethereal > > to decode the packet in a number of way. So should I decode it at the link layer as MPLS packet to see the header? > > > > Thanks again > > Amrit > > > > -------------------------------- > > Amritpal Singh > > Infosys Technologies Ltd. > > Bangalore - 561229 > > Tel No : 4166308 > > --------------------------------- > > > -- > James R. Leu > > > --__--__-- > > Message: 2 > Date: Fri, 18 Oct 2002 12:52:16 -0500 > From: "James R. Leu" <jl...@mi...> > To: =?iso-8859-1?Q?Fran=E7ois_Desloges?= <fde...@sy...> > Cc: mpl...@li... > Subject: Re: [mpls-linux-general] make problem with mplsadm on fresh Mdk9.0 distro > Reply-To: jl...@mi... > Organization: none > > On Thu, Oct 17, 2002 at 04:31:47PM -0400, Fran=E7ois Desloges wrote: > > Hi Jim et all, > >=20 > > I've build 2.4.19+ mpls-linux-1.1 on a fresh Mandrake 9.0 distro. > > I've rebooted and can now see the /proc/net/mpls_* (empty) entries > > I've changed the /usr/include/linux and asm link t use my new mpls kern= > el=20 > > include directory. > >=20 > > Doing make in utils results in: > >=20 > > gcc -g -Wall -c -o mplsadm.o mplsadm.c > > gcc -g -static -o mplsadm2 mplsadm.o > > /usr/bin/ld: cannot find -lc > > collect2: ld returned 1 exit status > > make: *** [mplsadm2] Error 1 > >=20 > > The mdk 9.0 stock gcc and ld version are > > gcc version 3.2 (Mandrake Linux 9.0 3.2-1mdk) > > GNU ld version 2.12.90.0.15 20020717 > > Argh. I haven't made the move to gcc 3.x yet. Does the lates debain hav= > e > gcc 3.x yet? If so maybe I'll try moving one of my machine to gcc 3.x > (and assocaited binutils versions) and see if I can work out any of these > type of problems. > > You might try and remove the '-static' from the Makefile line, or > try and compile/link my hand and see if you can figure out a combination > that works. > > Jim > > >=20 > > man ld says: > > -larchive > > --library=3Darchive > > Add archive file archive to the list of files to link. This= > option > > may be used any number of times. ld will search its path-l= > ist for > > occurrences of "libarchive.a" for every archive specified. > >=20 > > On systems which support shared libraries, ld may also sear= > ch for > > libraries with extensions other than ".a". Specifically,= > on ELF > > and SunOS systems, ld will search a directory for a library = > with an > > extension of ".so" before searching for one with an exten= > sion of > > ".a". By convention, a ".so" extension indicates a shared l= > ibrary. > >=20 > > The linker will search an archive only once, at the locatio= > n where > > it is specified on the command line. If the archive defines= > a sym- > > bol which was undefined in some object which appeared bef= > ore the > > archive on the command line, the linker will include the ap= > propri- > > ate file(s) from the archive. However, an undefined symbo= > l in an > > object appearing later on the command line will not cau= > se the > > linker to search the archive again. > >=20 > > See the -( option for a way to force the linker to search a= > rchives > > multiple times. > >=20 > > Where am I suppose to find the "c" "archive" (I obviously know nothing = > about=20 > > ld) ? > >=20 > > Thanks! > >=20 > > FD=20 > >=20 > >=20 > >=20 > > ------------------------------------------------------- > > This sf.net email is sponsored by: viaVerio will pay you up to > > $1,000 for every account that you consolidate with us. > > http://ad.doubleclick.net/clk;4749864;7604308;v? > > http://www.viaverio.com/consolidator/osdn.cfm > > _______________________________________________ > > mpls-linux-general mailing list > > mpl...@li... > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > --=20 > James R. Leu > > > --__--__-- > > Message: 3 > Date: Fri, 18 Oct 2002 12:55:07 -0500 > From: "James R. Leu" <jl...@mi...> > To: Alexandre Carlos <ale...@re...> > Cc: mpl...@li... > Subject: Re: [mpls-linux-general] Re: Dynamic LSP > Reply-To: jl...@mi... > Organization: none > > Before turning on LDP, you should make sure that OSPF has propogated the > routes for the FECs you want to create LSPs for. Also, make sure to > use the version of the zebra-ldp patch that I will be posting tonight. > > Jim > > On Thu, Oct 17, 2002 at 11:24:25PM -0300, Alexandre Carlos wrote: > > I configures my hosts to create a dynamic lsp, but they r not swaping > > labels, how can i configure them to do that? Or they should do that > > automaticly? > > > > Help would be very thank full > > > > Alex > > > > > > > > > > ------------------------------------------------------- > > This sf.net email is sponsored by: viaVerio will pay you up to > > $1,000 for every account that you consolidate with us. > > http://ad.doubleclick.net/clk;4749864;7604308;v? > > http://www.viaverio.com/consolidator/osdn.cfm > > _______________________________________________ > > mpls-linux-general mailing list > > mpl...@li... > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > -- > James R. Leu > > > --__--__-- > > Message: 4 > Date: Fri, 18 Oct 2002 14:17:08 -0500 > From: "James R. Leu" <jl...@mi...> > To: Sameer Gharat <sam...@ya...> > Cc: mpl...@so... > Subject: Re: [mpls-linux-general] LDP config... confused! > Reply-To: jl...@mi... > Organization: none > > Make sure you clean out all interface address and routes before starting > zebra/ospfd. Zebra treats interface that it provisioned and interfaces > learned from the kernel differntly. I do not understand all of > the ramification of this yet, but I work around it by configuring > everything via zebra. > > On Tue, Oct 15, 2002 at 06:20:16PM +0100, Sameer Gharat wrote: > > Hi James, > > > > nope... I have only ospfd running as the routing > > protocol. the funny thing is that mplsd does create > > the outlabels sometimes while sometimes it doesn't! > > ... I haven't been able to pin-point the conditions > > that exist when the out-labels are getting created. > > Maybe I'll go through the trace output and see if I > > can figure out something. > > But as far as existing route entry is concerned, I > > believe it would have been created by ospfd. Is that > > right? > > > > thanks and regards, > > Sameer > > > > --- "James R. Leu" <jl...@mi...> wrote: > Do > > you have some other routing protocol running? > > > LDP out labels only get > > > installed for FECs that match an existing route > > > entry. > > > > > > On Mon, Oct 14, 2002 at 04:01:00PM +0100, Sameer > > > Gharat wrote: > > > > Hello James and all, > > > > > > > > I included a dummy interface with a /32 address > > > and > > > > tried the procedure over again. It worked!! ... > > > though > > > > only partially, as it created incoming labels on > > > both > > > > the hosts that I was working on. But it did not > > > create > > > > any outgoing labels. > > > > > > > > Any reason, why this should happen? > > > > > > > > thanks in advance! > > > > > > > > cheers, > > > > Sameer > > > > > > > > --- "James R. Leu" <jl...@mi...> wrote: > > > > The > > > > default behavior for zebra-ldp is to only > > > > > distribute labels for > > > > > the LSR-ID (this can be changed, but only > > > > > programitically). The LSR-ID > > > > > must be attached to a directly connected > > > interface > > > > > and must show up as > > > > > a /32 in the routing table (this is probably a > > > > > little too strict, I need to > > > > > tweak these rules). So the easiest way is to > > > use a > > > > > 'dummy' interface. > > > > > Compile your kernel with support for it, then > > > assign > > > > > a /32 to it through zebra. > > > > > Configure OSPF to run passive on the dummy > > > > > interface. You should then see > > > > > that LDP distruibutes labels for the /32s. > > > > > > > > > > Jim > > > > > > > > > > > > > > > > > ________________________________________________________________________ > > > > Missed your favourite TV serial last night? Try > > > the new, Yahoo! TV. > > > > visit http://in.tv.yahoo.com > > > > > > > > > > > > > > > > > ------------------------------------------------------- > > > > This sf.net email is sponsored by:ThinkGeek > > > > Welcome to geek heaven. > > > > http://thinkgeek.com/sf > > > > _______________________________________________ > > > > mpls-linux-general mailing list > > > > mpl...@li... > > > > > > > > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > > > > > -- > > > James R. Leu > > > > ________________________________________________________________________ > > Missed your favourite TV serial last night? Try the new, Yahoo! TV. > > visit http://in.tv.yahoo.com > > > > > > ------------------------------------------------------- > > This sf.net email is sponsored by:ThinkGeek > > Welcome to geek heaven. > > http://thinkgeek.com/sf > > _______________________________________________ > > mpls-linux-general mailing list > > mpl...@li... > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > -- > James R. Leu > > > --__--__-- > > Message: 5 > Date: Fri, 18 Oct 2002 14:25:30 -0500 > From: "James R. Leu" <jl...@mi...> > To: Anjaneya Pal <anj...@ya...> > Cc: mpl...@li... > Subject: Re: [mpls-linux-general] ldp with ip unnumbered > Reply-To: jl...@mi... > Organization: none > > On Sat, Oct 05, 2002 at 07:54:21PM +0530, Anjaneya Pal wrote: > > hi , > > the current ldp make it imperative for all the interfaces in a router to have > > an unique ip address. can we have an ldp which can even run even if the > > interfaces are unnumbered....as in ospf. Thus we would not waste ip addresses > > for point-to-point links. > > I know the way LDP is supposed to handle un-numbered interfaces, I just have > not had time to implement/test it in ldp-portable. As an aside, not many > service providers actually use un-numbered interfaces, it is at most and > interesting exercise. > > Jim > > > > > thnx in advance > > anjaneya > > > > > > ------------------------------------------------------- > > This sf.net email is sponsored by:ThinkGeek > > Welcome to geek heaven. > > http://thinkgeek.com/sf > > _______________________________________________ > > mpls-linux-general mailing list > > mpl...@li... > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > -- > James R. Leu > > > > --__--__-- > > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > > End of mpls-linux-general Digest --------------------------------------------- This message was sent using sharif web-based mail. http://mehr.sharif.edu |
|
From: <sam...@ya...> - 2002-10-19 08:57:18
|
Hello James,
Yup!... I did clean out all the interface addresses
and routes and made a clean start.
And after a bit of struggle, I managed to get it to
work!! ... though it did take time for me to realize
the significance of "backbone area" in ospfd. Once
that issue was ironed out, LDP started working like a
dream and exchanged labels and created dynamic LSPs
perfectly!
Thanks a ton for all the help you provided from time
to time as I hit some raodblocks along the way!!
Thanks a lot!
cheers,
Sameer
--- "James R. Leu" <jl...@mi...> wrote: >
Make sure you clean out all interface address and
> routes before starting
> zebra/ospfd. Zebra treats interface that it
> provisioned and interfaces
> learned from the kernel differntly. I do not
> understand all of
> the ramification of this yet, but I work around it
> by configuring
> everything via zebra.
>
> On Tue, Oct 15, 2002 at 06:20:16PM +0100, Sameer
> Gharat wrote:
________________________________________________________________________
Missed your favourite TV serial last night? Try the new, Yahoo! TV.
visit http://in.tv.yahoo.com
|
|
From: Folke A. <aeo...@ho...> - 2002-10-19 07:52:12
|
hi, all
will anyone tell me the meaning of the following parameters in
/proc/net/mpls* :?
[root@localhost net]# cat mpls_in
0x40004000 4/256/0 gen 16 0 1 POP PEEK
+ + +
| | |
| | +------what does this
| | mean then?
| |
| label space?
|
what does this zero mean? packets
dropped? or bytest dropped?
[root@localhost net]# cat mpls_out
0x00000002 7837/477621/0 3 PUSH(gen 230) SET(eth1,11.0.2.2)
+
|
|
what does this mean?
[root@localhost net]# cat mpls_labelspace
eth0 0 11
+
|
|
and what about this?
thanks in advacne
folke.
_________________________________________________________________
Internet access plans that fit your lifestyle -- join MSN.
http://resourcecenter.msn.com/access/plans/default.asp
|
|
From: James R. L. <jl...@mi...> - 2002-10-19 03:23:03
|
At one point I had entertained the idea of distributing binaries, -static is a good way to get around any glibc version mismatches. Jim On Fri, Oct 18, 2002 at 04:56:33PM -0400, Fran=E7ois Desloges wrote: > Removing the -static worked. > Having installed the mandrake glibc-static-devel package would have bee= n a=20 > good idea as well ;-) >=20 > By the way is there a good reason to link it static ? > (Isn't glibc.so always there?) >=20 > FD >=20 > On October 18, 2002 13:52, James R. Leu wrote: > > On Thu, Oct 17, 2002 at 04:31:47PM -0400, Fran=E7ois Desloges wrote: > > > Hi Jim et all, > > > > > > I've build 2.4.19+ mpls-linux-1.1 on a fresh Mandrake 9.0 distro. > > > I've rebooted and can now see the /proc/net/mpls_* (empty) entries > > > I've changed the /usr/include/linux and asm link t use my new mpls = kernel > > > include directory. > > > > > > Doing make in utils results in: > > > > > > gcc -g -Wall -c -o mplsadm.o mplsadm.c > > > gcc -g -static -o mplsadm2 mplsadm.o > > > /usr/bin/ld: cannot find -lc > > > collect2: ld returned 1 exit status > > > make: *** [mplsadm2] Error 1 > > > > > > The mdk 9.0 stock gcc and ld version are > > > gcc version 3.2 (Mandrake Linux 9.0 3.2-1mdk) > > > GNU ld version 2.12.90.0.15 20020717 > > > > Argh. I haven't made the move to gcc 3.x yet. Does the lates debain= have > > gcc 3.x yet? If so maybe I'll try moving one of my machine to gcc 3.= x > > (and assocaited binutils versions) and see if I can work out any of t= hese > > type of problems. > > > > You might try and remove the '-static' from the Makefile line, or > > try and compile/link my hand and see if you can figure out a combinat= ion > > that works. > > > > Jim > > > > > man ld says: > > > -larchive > > > --library=3Darchive > > > Add archive file archive to the list of files to link. = This > > > option may be used any number of times. ld will search its path-l= ist > > > for occurrences of "libarchive.a" for every archive specified. > > > > > > On systems which support shared libraries, ld may also = search > > > for libraries with extensions other than ".a". Specifically, o= n ELF > > > and SunOS systems, ld will search a directory for a library with an > > > extension of ".so" before searching for one with an extension of= ".a". > > > By convention, a ".so" extension indicates a shared library. > > > > > > The linker will search an archive only once, at the loc= ation > > > where it is specified on the command line. If the archive defines = a sym- > > > bol which was undefined in some object which appeared before the > > > archive on the command line, the linker will include the appropri-= ate=20 > > > file(s) from the archive. However, an undefined symbol in an obje= ct > > > appearing later on the command line will not cause the linker= to > > > search the archive again. > > > > > > See the -( option for a way to force the linker to sear= ch > > > archives multiple times. > > > > > > Where am I suppose to find the "c" "archive" (I obviously know noth= ing > > > about ld) ? > > > > > > Thanks! > > > > > > FD > > > > > > > > > > > > ------------------------------------------------------- > > > This sf.net email is sponsored by: viaVerio will pay you up to > > > $1,000 for every account that you consolidate with us. > > > http://ad.doubleclick.net/clk;4749864;7604308;v? > > > http://www.viaverio.com/consolidator/osdn.cfm > > > _______________________________________________ > > > mpls-linux-general mailing list > > > mpl...@li... > > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general >=20 >=20 >=20 > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu |
|
From: <fde...@sy...> - 2002-10-18 21:01:17
|
Removing the -static worked. Having installed the mandrake glibc-static-devel package would have been a good idea as well ;-) By the way is there a good reason to link it static ? (Isn't glibc.so always there?) FD On October 18, 2002 13:52, James R. Leu wrote: > On Thu, Oct 17, 2002 at 04:31:47PM -0400, François Desloges wrote: > > Hi Jim et all, > > > > I've build 2.4.19+ mpls-linux-1.1 on a fresh Mandrake 9.0 distro. > > I've rebooted and can now see the /proc/net/mpls_* (empty) entries > > I've changed the /usr/include/linux and asm link t use my new mpls kernel > > include directory. > > > > Doing make in utils results in: > > > > gcc -g -Wall -c -o mplsadm.o mplsadm.c > > gcc -g -static -o mplsadm2 mplsadm.o > > /usr/bin/ld: cannot find -lc > > collect2: ld returned 1 exit status > > make: *** [mplsadm2] Error 1 > > > > The mdk 9.0 stock gcc and ld version are > > gcc version 3.2 (Mandrake Linux 9.0 3.2-1mdk) > > GNU ld version 2.12.90.0.15 20020717 > > Argh. I haven't made the move to gcc 3.x yet. Does the lates debain have > gcc 3.x yet? If so maybe I'll try moving one of my machine to gcc 3.x > (and assocaited binutils versions) and see if I can work out any of these > type of problems. > > You might try and remove the '-static' from the Makefile line, or > try and compile/link my hand and see if you can figure out a combination > that works. > > Jim > > > man ld says: > > -larchive > > --library=archive > > Add archive file archive to the list of files to link. This > > option may be used any number of times. ld will search its path-list > > for occurrences of "libarchive.a" for every archive specified. > > > > On systems which support shared libraries, ld may also search > > for libraries with extensions other than ".a". Specifically, on ELF > > and SunOS systems, ld will search a directory for a library with an > > extension of ".so" before searching for one with an extension of ".a". > > By convention, a ".so" extension indicates a shared library. > > > > The linker will search an archive only once, at the location > > where it is specified on the command line. If the archive defines a sym- > > bol which was undefined in some object which appeared before the > > archive on the command line, the linker will include the appropri- ate > > file(s) from the archive. However, an undefined symbol in an object > > appearing later on the command line will not cause the linker to > > search the archive again. > > > > See the -( option for a way to force the linker to search > > archives multiple times. > > > > Where am I suppose to find the "c" "archive" (I obviously know nothing > > about ld) ? > > > > Thanks! > > > > FD > > > > > > > > ------------------------------------------------------- > > This sf.net email is sponsored by: viaVerio will pay you up to > > $1,000 for every account that you consolidate with us. > > http://ad.doubleclick.net/clk;4749864;7604308;v? > > http://www.viaverio.com/consolidator/osdn.cfm > > _______________________________________________ > > mpls-linux-general mailing list > > mpl...@li... > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general |
|
From: James R. L. <jl...@mi...> - 2002-10-18 18:26:07
|
On Sat, Oct 05, 2002 at 07:54:21PM +0530, Anjaneya Pal wrote: > hi , > the current ldp make it imperative for all the interfaces in a router to have > an unique ip address. can we have an ldp which can even run even if the > interfaces are unnumbered....as in ospf. Thus we would not waste ip addresses > for point-to-point links. I know the way LDP is supposed to handle un-numbered interfaces, I just have not had time to implement/test it in ldp-portable. As an aside, not many service providers actually use un-numbered interfaces, it is at most and interesting exercise. Jim > > thnx in advance > anjaneya > > > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu |
|
From: James R. L. <jl...@mi...> - 2002-10-18 18:15:39
|
Make sure you clean out all interface address and routes before starting zebra/ospfd. Zebra treats interface that it provisioned and interfaces learned from the kernel differntly. I do not understand all of the ramification of this yet, but I work around it by configuring everything via zebra. On Tue, Oct 15, 2002 at 06:20:16PM +0100, Sameer Gharat wrote: > Hi James, > > nope... I have only ospfd running as the routing > protocol. the funny thing is that mplsd does create > the outlabels sometimes while sometimes it doesn't! > ... I haven't been able to pin-point the conditions > that exist when the out-labels are getting created. > Maybe I'll go through the trace output and see if I > can figure out something. > But as far as existing route entry is concerned, I > believe it would have been created by ospfd. Is that > right? > > thanks and regards, > Sameer > > --- "James R. Leu" <jl...@mi...> wrote: > Do > you have some other routing protocol running? > > LDP out labels only get > > installed for FECs that match an existing route > > entry. > > > > On Mon, Oct 14, 2002 at 04:01:00PM +0100, Sameer > > Gharat wrote: > > > Hello James and all, > > > > > > I included a dummy interface with a /32 address > > and > > > tried the procedure over again. It worked!! ... > > though > > > only partially, as it created incoming labels on > > both > > > the hosts that I was working on. But it did not > > create > > > any outgoing labels. > > > > > > Any reason, why this should happen? > > > > > > thanks in advance! > > > > > > cheers, > > > Sameer > > > > > > --- "James R. Leu" <jl...@mi...> wrote: > > > The > > > default behavior for zebra-ldp is to only > > > > distribute labels for > > > > the LSR-ID (this can be changed, but only > > > > programitically). The LSR-ID > > > > must be attached to a directly connected > > interface > > > > and must show up as > > > > a /32 in the routing table (this is probably a > > > > little too strict, I need to > > > > tweak these rules). So the easiest way is to > > use a > > > > 'dummy' interface. > > > > Compile your kernel with support for it, then > > assign > > > > a /32 to it through zebra. > > > > Configure OSPF to run passive on the dummy > > > > interface. You should then see > > > > that LDP distruibutes labels for the /32s. > > > > > > > > Jim > > > > > > > > > > > > ________________________________________________________________________ > > > Missed your favourite TV serial last night? Try > > the new, Yahoo! TV. > > > visit http://in.tv.yahoo.com > > > > > > > > > > > > ------------------------------------------------------- > > > This sf.net email is sponsored by:ThinkGeek > > > Welcome to geek heaven. > > > http://thinkgeek.com/sf > > > _______________________________________________ > > > mpls-linux-general mailing list > > > mpl...@li... > > > > > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > > > -- > > James R. Leu > > ________________________________________________________________________ > Missed your favourite TV serial last night? Try the new, Yahoo! TV. > visit http://in.tv.yahoo.com > > > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu |
|
From: James R. L. <jl...@mi...> - 2002-10-18 16:53:52
|
Before turning on LDP, you should make sure that OSPF has propogated the routes for the FECs you want to create LSPs for. Also, make sure to use the version of the zebra-ldp patch that I will be posting tonight. Jim On Thu, Oct 17, 2002 at 11:24:25PM -0300, Alexandre Carlos wrote: > I configures my hosts to create a dynamic lsp, but they r not swaping > labels, how can i configure them to do that? Or they should do that > automaticly? > > Help would be very thank full > > Alex > > > > > ------------------------------------------------------- > This sf.net email is sponsored by: viaVerio will pay you up to > $1,000 for every account that you consolidate with us. > http://ad.doubleclick.net/clk;4749864;7604308;v? > http://www.viaverio.com/consolidator/osdn.cfm > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu |
|
From: James R. L. <jl...@mi...> - 2002-10-18 16:50:48
|
On Thu, Oct 17, 2002 at 04:31:47PM -0400, Fran=E7ois Desloges wrote: > Hi Jim et all, >=20 > I've build 2.4.19+ mpls-linux-1.1 on a fresh Mandrake 9.0 distro. > I've rebooted and can now see the /proc/net/mpls_* (empty) entries > I've changed the /usr/include/linux and asm link t use my new mpls kern= el=20 > include directory. >=20 > Doing make in utils results in: >=20 > gcc -g -Wall -c -o mplsadm.o mplsadm.c > gcc -g -static -o mplsadm2 mplsadm.o > /usr/bin/ld: cannot find -lc > collect2: ld returned 1 exit status > make: *** [mplsadm2] Error 1 >=20 > The mdk 9.0 stock gcc and ld version are > gcc version 3.2 (Mandrake Linux 9.0 3.2-1mdk) > GNU ld version 2.12.90.0.15 20020717 Argh. I haven't made the move to gcc 3.x yet. Does the lates debain hav= e gcc 3.x yet? If so maybe I'll try moving one of my machine to gcc 3.x (and assocaited binutils versions) and see if I can work out any of these type of problems. You might try and remove the '-static' from the Makefile line, or try and compile/link my hand and see if you can figure out a combination that works. Jim >=20 > man ld says: > -larchive > --library=3Darchive > Add archive file archive to the list of files to link. This= option > may be used any number of times. ld will search its path-l= ist for > occurrences of "libarchive.a" for every archive specified. >=20 > On systems which support shared libraries, ld may also sear= ch for > libraries with extensions other than ".a". Specifically,= on ELF > and SunOS systems, ld will search a directory for a library = with an > extension of ".so" before searching for one with an exten= sion of > ".a". By convention, a ".so" extension indicates a shared l= ibrary. >=20 > The linker will search an archive only once, at the locatio= n where > it is specified on the command line. If the archive defines= a sym- > bol which was undefined in some object which appeared bef= ore the > archive on the command line, the linker will include the ap= propri- > ate file(s) from the archive. However, an undefined symbo= l in an > object appearing later on the command line will not cau= se the > linker to search the archive again. >=20 > See the -( option for a way to force the linker to search a= rchives > multiple times. >=20 > Where am I suppose to find the "c" "archive" (I obviously know nothing = about=20 > ld) ? >=20 > Thanks! >=20 > FD=20 >=20 >=20 >=20 > ------------------------------------------------------- > This sf.net email is sponsored by: viaVerio will pay you up to > $1,000 for every account that you consolidate with us. > http://ad.doubleclick.net/clk;4749864;7604308;v? > http://www.viaverio.com/consolidator/osdn.cfm > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general --=20 James R. Leu |
|
From: James R. L. <jl...@mi...> - 2002-10-18 16:43:35
|
The first change I suggest you make is to switch to the latest version of mpls-linux-1.1. Many things have changed since 1.0, and I do not have the time to downgrade to mpls-linux-1.0 to test/verify any of the mpls-linux-1.0 command you are try, but...... ... since it seems like you need to stay with mpls-linux-1.0 the only thing I can see wrong (from a quick skim of your output) is that when specifing -f <prefix>, the prefix must have an EXACT match in your routing table. In your example below, 10.2.0.2/32 must show up in your routing table. The "mpls_bind_out2fec: couldn't find the FIB_NODE(0)" message in the 'dmesg' output is what tells me that 10.2.0.2/32 is not in your routing table. Jim On Fri, Oct 18, 2002 at 06:20:46PM +0530, Amritpal Singh wrote: > > Hi All, > > The following is the setup in which I am trying to test MPLS. > > > -------- ----------------- ----------------- --------- > | |------------------ | |------------------| |------------------- | | > | A | | B | | C | | D | > -------- ----------------- ----------------- --------- > 10.1.1.2 10.1.1.1 10.2.0.1 10.2.0.2 10.2.1.1 10.2.1.10 > eth0 eth0 eth1 eth0 eth1 eth0 > > I am running mpls-linux-1.0 on a Red Hat Linux 7.2 > All the nodes hav the patch applied to them. > > Nodes B and C are the ones which should enable me to see label switching. > So I configure the incoming and outgoing labels on B and C as follows. > > **************FOR NODE B ****************************************************** > [root@ferrari utils]# ./mplsadm -v -A -O gen:16:eth1:ipv4:10.2.0.2 > > Out label input: gen:16:eth1:ipv4:10.2.0.2 > > ipv4 > > [root@ferrari utils]# > > [root@ferrari utils]# ./mplsadm -B -O gen:16:eth1 -f 10.2.0.2/32 > > [root@ferrari utils]# > > [root@ferrari utils]# ./mplsadm -L eth1:0 > > [root@ferrari utils]# ./mplsadm -v -A -I gen:17:0 > > In label input: gen:17:0 > > [root@ferrari utils]# ./mplsadm -d > > Debug: Success > > [root@ferrari utils]# ./mplsadm -v -B -O gen:16:eth1 -f 10.2.0.2/32 > > Out label input: gen:16:eth1 > > FEC input: 10.2.0.2/32 > > ****************END FOR NODE B *********************************************** > > ****************FOR NODE C ***************************************************** > [root@renault utils]# ./mplsadm -v -A -O gen:17:eth0:ipv4:10.2.0.1 > > Out label input: gen:17:eth0:ipv4:10.2.0.1 > > ipv4 > > [root@renault utils]# > > [root@renault utils]# ./mplsadm -v -B -O gen:17:eth0 -f 10.2.0.1/32 > > Out label input: gen:17:eth0 > > FEC input: 10.2.0.1/32 > > [root@renault utils]# ./mplsadm -L eth0:0 > > [root@renault utils]# ./mplsadm -v -A -I gen:16:0 > > In label input: gen:16:0 > > [root@renault utils]# ./mplsadm -L eth1:0 > > [root@renault utils]# > > *****************************END FOR NODE C ************************************* > > > > I see all the valid entries in all the mpls_* files in /proc/net as shown below. > > **********FOR NODE B ************************************************************** > [root@ferrari Desktop]# more /proc/net/mpls_* > > :::::::::::::: > > /proc/net/mpls_fec > > :::::::::::::: > > :::::::::::::: > > /proc/net/mpls_in > > :::::::::::::: > > 40004400 0/0/0 gen 17 0 POP PEEK > > :::::::::::::: > > /proc/net/mpls_labelspace > > :::::::::::::: > > lo 0 > > eth1 0 > > :::::::::::::: > > /proc/net/mpls_out > > :::::::::::::: > > 40004003 0/0/0 PUSH(gen 16) SET(eth1,10.2.0.2) > > :::::::::::::: > > /proc/net/mpls_tunnel > > :::::::::::::: > > *************END FOR NODE B *************************************************** > ***************FOR NODE C ****************************************************** > [root@renault root]# more /proc/net/mpls_* > > :::::::::::::: > > /proc/net/mpls_fec > > :::::::::::::: > > :::::::::::::: > > /proc/net/mpls_in > > :::::::::::::: > > 40004000 0/0/0 gen 16 0 POP PEEK > > :::::::::::::: > > /proc/net/mpls_labelspace > > :::::::::::::: > > lo 0 > > eth0 0 > > :::::::::::::: > > /proc/net/mpls_out > > :::::::::::::: > > 40004403 0/0/0 PUSH(gen 17) SET(eth0,10.2.0.1) > > :::::::::::::: > > /proc/net/mpls_tunnel > > :::::::::::::: > > > > But when I try to ping D from A I donot see the label switching happening at all.Ping works fine. > > But the counters in mpls_in and mpls_out remain 0.Plus I donot see anything in mpls_fec.Should I ? > > I run ethereal on B and C to see incoming/outgoing labels on the packet but I find that > there is absolutely no label on the packets. > > The last few lines of output of "DMESG" are shown below . > ************************FOR NODE B ***************************************** > MPLS version 0.996 11/21/2001 jl...@mi... > > MPLS Tunnel interface > > ds: no socket drivers loaded! > > VFS: Mounted root (ext2 filesystem) readonly. > > Freeing unused kernel memory: 224k freed > > Adding Swap: 153176k swap-space (priority -1) > > mpls_prep_out2fec: enter > > Label GEN 16 > > Key GEN 16 3 > > mpls_fill_key_res: enter > > mpls_fill_key_res: exit > > mpls_bind_out2fec: couldn't find the FIB_NODE(0) > > mpls_bind_out2fec: exit > > mpls_set_labelspace: enter > > mpls_set_labelspace: labelspace(0) > > mpls_set_labelspace: exit > > *****************************END FOR NODE B ******************************* > ***************************** FOR NODE C ************************************ > MPLS version 0.996 11/21/2001 jl...@mi... > > MPLS Tunnel interface > > ds: no socket drivers loaded! > > VFS: Mounted root (ext2 filesystem) readonly. > > Freeing unused kernel memory: 224k freed > > Adding Swap: 153176k swap-space (priority -1) > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > device eth0 entered promiscuous mode > > device eth0 left promiscuous mode > > *******************************************END FOR NODE C ******************************** > > I cannot upgrade to the latest version available because of time constraints.I had downloaded this > version some months back and I have to show MPLS working with this version only. > > I will be very thankful to all the people who respond and help me in any way. > > I am looking forward to your comments on this. > > P.S: Last query: Do I need to "decode" the ICMP packet received at node C by ethereal into a MPLS packet. > As I see the raw packet received at the ethereal there is no MPLS header. But there is an option in ethereal > to decode the packet in a number of way. So should I decode it at the link layer as MPLS packet to see the header? > > Thanks again > Amrit > > -------------------------------- > Amritpal Singh > Infosys Technologies Ltd. > Bangalore - 561229 > Tel No : 4166308 > --------------------------------- -- James R. Leu |
|
From: Alexandre C. <ale...@re...> - 2002-10-18 02:22:35
|
I configures my hosts to create a dynamic lsp, but they r not swaping labels, how can i configure them to do that? Or they should do that automaticly? Help would be very thank full Alex |
|
From: <fde...@sy...> - 2002-10-17 20:36:29
|
Hi Jim et all,
I've build 2.4.19+ mpls-linux-1.1 on a fresh Mandrake 9.0 distro.
I've rebooted and can now see the /proc/net/mpls_* (empty) entries
I've changed the /usr/include/linux and asm link t use my new mpls kernel
include directory.
Doing make in utils results in:
gcc -g -Wall -c -o mplsadm.o mplsadm.c
gcc -g -static -o mplsadm2 mplsadm.o
/usr/bin/ld: cannot find -lc
collect2: ld returned 1 exit status
make: *** [mplsadm2] Error 1
The mdk 9.0 stock gcc and ld version are
gcc version 3.2 (Mandrake Linux 9.0 3.2-1mdk)
GNU ld version 2.12.90.0.15 20020717
man ld says:
-larchive
--library=archive
Add archive file archive to the list of files to link. This option
may be used any number of times. ld will search its path-list for
occurrences of "libarchive.a" for every archive specified.
On systems which support shared libraries, ld may also search for
libraries with extensions other than ".a". Specifically, on ELF
and SunOS systems, ld will search a directory for a library with an
extension of ".so" before searching for one with an extension of
".a". By convention, a ".so" extension indicates a shared library.
The linker will search an archive only once, at the location where
it is specified on the command line. If the archive defines a sym-
bol which was undefined in some object which appeared before the
archive on the command line, the linker will include the appropri-
ate file(s) from the archive. However, an undefined symbol in an
object appearing later on the command line will not cause the
linker to search the archive again.
See the -( option for a way to force the linker to search archives
multiple times.
Where am I suppose to find the "c" "archive" (I obviously know nothing about
ld) ?
Thanks!
FD
|
|
From: James R. L. <jl...@mi...> - 2002-10-17 12:26:35
|
What version of ldp-portable and zebra are you using? On Wed, Oct 16, 2002 at 10:50:34PM -0700, Paresh Patil wrote: > Hi, > > I am trying to run MPLS-LDP on two machines connected > by a cross-cable. > > Each machine has two cards, > other card of machine A is connected to a subnet and > that of Mchine B is connected to different subnet. > Both the addresses are /24 addresses. > So I created a dummy interface on both the m/cs with > /32 address. > > After rebboting both the the machines in MPLS, I > configure zebra oboth the machines followed by > configuration of ospf. > > OSPF detects neighbors etc well and detects all three > intergaces etc. This goes well. > > After this I start mpls and configures it. > > It just generates inlabel and no outlabel. I don't > know why is it doing so? When I repeated this with > some-change (must have assigned different ips or > interchanged eth0 to eth1.. something I don't > remember) it generated both the labels. But never > again after that. Why is it not generating outlabel? > > Regards, > Paresh > > > > __________________________________________________ > Do you Yahoo!? > Faith Hill - Exclusive Performances, Videos & More > http://faith.yahoo.com > > > ------------------------------------------------------- > This sf.net email is sponsored by: viaVerio will pay you up to > $1,000 for every account that you consolidate with us. > http://ad.doubleclick.net/clk;4749864;7604308;v? > http://www.viaverio.com/consolidator/osdn.cfm > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu |
|
From: Paresh P. <par...@ya...> - 2002-10-17 05:50:35
|
Hi, I am trying to run MPLS-LDP on two machines connected by a cross-cable. Each machine has two cards, other card of machine A is connected to a subnet and that of Mchine B is connected to different subnet. Both the addresses are /24 addresses. So I created a dummy interface on both the m/cs with /32 address. After rebboting both the the machines in MPLS, I configure zebra oboth the machines followed by configuration of ospf. OSPF detects neighbors etc well and detects all three intergaces etc. This goes well. After this I start mpls and configures it. It just generates inlabel and no outlabel. I don't know why is it doing so? When I repeated this with some-change (must have assigned different ips or interchanged eth0 to eth1.. something I don't remember) it generated both the labels. But never again after that. Why is it not generating outlabel? Regards, Paresh __________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com |
|
From: <sam...@ya...> - 2002-10-15 17:20:25
|
Hi James, nope... I have only ospfd running as the routing protocol. the funny thing is that mplsd does create the outlabels sometimes while sometimes it doesn't! ... I haven't been able to pin-point the conditions that exist when the out-labels are getting created. Maybe I'll go through the trace output and see if I can figure out something. But as far as existing route entry is concerned, I believe it would have been created by ospfd. Is that right? thanks and regards, Sameer --- "James R. Leu" <jl...@mi...> wrote: > Do you have some other routing protocol running? > LDP out labels only get > installed for FECs that match an existing route > entry. > > On Mon, Oct 14, 2002 at 04:01:00PM +0100, Sameer > Gharat wrote: > > Hello James and all, > > > > I included a dummy interface with a /32 address > and > > tried the procedure over again. It worked!! ... > though > > only partially, as it created incoming labels on > both > > the hosts that I was working on. But it did not > create > > any outgoing labels. > > > > Any reason, why this should happen? > > > > thanks in advance! > > > > cheers, > > Sameer > > > > --- "James R. Leu" <jl...@mi...> wrote: > > The > > default behavior for zebra-ldp is to only > > > distribute labels for > > > the LSR-ID (this can be changed, but only > > > programitically). The LSR-ID > > > must be attached to a directly connected > interface > > > and must show up as > > > a /32 in the routing table (this is probably a > > > little too strict, I need to > > > tweak these rules). So the easiest way is to > use a > > > 'dummy' interface. > > > Compile your kernel with support for it, then > assign > > > a /32 to it through zebra. > > > Configure OSPF to run passive on the dummy > > > interface. You should then see > > > that LDP distruibutes labels for the /32s. > > > > > > Jim > > > > > > > ________________________________________________________________________ > > Missed your favourite TV serial last night? Try > the new, Yahoo! TV. > > visit http://in.tv.yahoo.com > > > > > > > ------------------------------------------------------- > > This sf.net email is sponsored by:ThinkGeek > > Welcome to geek heaven. > > http://thinkgeek.com/sf > > _______________________________________________ > > mpls-linux-general mailing list > > mpl...@li... > > > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general > > -- > James R. Leu ________________________________________________________________________ Missed your favourite TV serial last night? Try the new, Yahoo! TV. visit http://in.tv.yahoo.com |
|
From: Pim V. H. <pim...@in...> - 2002-10-14 21:50:01
|
On Tue, 2002-10-08 at 11:33, Folke Aeon wrote: > hi, all :) >=20 > i got a problem in balancing my traffic through > my network. >=20 > see the following network topology: > +------+ +------+ +------+ | > | |eth1 | | | | |10.0.0.0/24 > | +------+ CORE +-----+ | +------- > eth0 | | | | | | | > ------+ LSR1 | +------+ | LSR2 +---+ > | |eth2 | | | > | +-------------------+ | > +------+ LINK0 +------+ >=20 > all three PCs are running kernel-2.4.19 with a > mpls patch 1.160. >=20 > on lsr1 , i have three ethether net adaptors. >=20 > eth0 is the ingress device > eth1 has an outgoing label 1 > eth2 has an outgoing label 2 >=20 > and two tunnels created by using 'mplsadm -T': >=20 > mpls12 : bind to eth2 > mpls1c : bind to eth1 >=20 > and its routing table contains the following > entry: >=20 > 10.0.0.0/24 dev mpls12 proto kernel > core dev mpls1c proto kernel >=20 > now , i want to make a flow with a certain > dst ip address , say , 10.0.0.111, to go throught > this network by balancing the traffic loading > onto eth1 and eth2 on demand. >=20 > e.g. when the direct link between lsr1 and > lsr2 (for convenience we named it LINK0 )are > lighted loaded, the traffic destined to > 11.0.0.111 goes out using mpls12 with a label 2. >=20 > when sometimes later, the loading on LINK0 > became heavier, i want to shift the traffic > destined to 11.0.0.111 to go out using mpls1c > with label 1. Well you need some kind of monitoring to determine the load. This is not really MPLS specific so you might want to search the net for a tool that fits your needs. Or you could write something yourself based on the statistics in /proc/net/mpls_*. Based on the monitoring information you can map/remap traffic on the different LSPs.=20 Another option would be to use some static load sharing. ip(8) supports adding multiple interfaces to a single route, the traffic is shared based on a hash on the IP header. Has anyone experience with that using MPLS? Pim. >=20 > can i achieve this ? > if i can , what should i do then ? >=20 > by the way , i am not using LDP currently. >=20 > thanks in advance. >=20 > Folke. >=20 > _________________________________________________________________ > Join the world=92s largest e-mail service with MSN Hotmail.=20 > http://www.hotmail.com >=20 >=20 >=20 > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general >=20 >=20 --=20 Pim Van Heuven Ghent University - IMEC Department of Information Technology (INTEC) Sint-Pietersnieuwstraat, 41 B-9000, Gent, Belgium pim...@in... DiffServ/MPLS(RSVP-TE) for Linux: http://dsmpls.atlantis.rug.ac.be |
|
From: James R. L. <jl...@mi...> - 2002-10-14 19:49:11
|
Do you have some other routing protocol running? LDP out labels only get installed for FECs that match an existing route entry. On Mon, Oct 14, 2002 at 04:01:00PM +0100, Sameer Gharat wrote: > Hello James and all, > > I included a dummy interface with a /32 address and > tried the procedure over again. It worked!! ... though > only partially, as it created incoming labels on both > the hosts that I was working on. But it did not create > any outgoing labels. > > Any reason, why this should happen? > > thanks in advance! > > cheers, > Sameer > > --- "James R. Leu" <jl...@mi...> wrote: > The > default behavior for zebra-ldp is to only > > distribute labels for > > the LSR-ID (this can be changed, but only > > programitically). The LSR-ID > > must be attached to a directly connected interface > > and must show up as > > a /32 in the routing table (this is probably a > > little too strict, I need to > > tweak these rules). So the easiest way is to use a > > 'dummy' interface. > > Compile your kernel with support for it, then assign > > a /32 to it through zebra. > > Configure OSPF to run passive on the dummy > > interface. You should then see > > that LDP distruibutes labels for the /32s. > > > > Jim > > > ________________________________________________________________________ > Missed your favourite TV serial last night? Try the new, Yahoo! TV. > visit http://in.tv.yahoo.com > > > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > mpls-linux-general mailing list > mpl...@li... > https://lists.sourceforge.net/lists/listinfo/mpls-linux-general -- James R. Leu |
|
From: <sam...@ya...> - 2002-10-14 15:01:19
|
Hello James and all,
I included a dummy interface with a /32 address and
tried the procedure over again. It worked!! ... though
only partially, as it created incoming labels on both
the hosts that I was working on. But it did not create
any outgoing labels.
Any reason, why this should happen?
thanks in advance!
cheers,
Sameer
--- "James R. Leu" <jl...@mi...> wrote: > The
default behavior for zebra-ldp is to only
> distribute labels for
> the LSR-ID (this can be changed, but only
> programitically). The LSR-ID
> must be attached to a directly connected interface
> and must show up as
> a /32 in the routing table (this is probably a
> little too strict, I need to
> tweak these rules). So the easiest way is to use a
> 'dummy' interface.
> Compile your kernel with support for it, then assign
> a /32 to it through zebra.
> Configure OSPF to run passive on the dummy
> interface. You should then see
> that LDP distruibutes labels for the /32s.
>
> Jim
________________________________________________________________________
Missed your favourite TV serial last night? Try the new, Yahoo! TV.
visit http://in.tv.yahoo.com
|
|
From: Alexandre C. <ale...@re...> - 2002-10-09 21:38:29
|
I=B4m having problem setting up the dynamic lsp. A have a network of 4 host two PE hosts two P hosts. First i started the zebra deamon, then the mpls deamon, configures it as = Jim told to, after that i started the ospf demand, on all host. After a while the telnet that i made to configure the mpls deamon freeze,= i think because of that the PE can=B4t discover any peers. Can anyone help me... Alex |
533 messages has been excluded from this view by a project administrator.
