Re: [mpls-linux-general] Multicast routing over linux mpls tunnel ?

[James R. L. <jl...@mi...>]

On Thu, Sep 09, 2004 at 03:11:43PM +1000, Jon...@cs... wrote:
> Hi James,
> =20
> First of all, thank you very much for your patch and comments...=20
> =20
> Before heading off this direction, I just want to ask if you have come ac=
ross a multicast enhancement of your code in=20
> =20
> http://www.cs.virginia.edu/~mngroup/projects/mpls/documents/thesis/node28=
.html

Yes.  I've talked with the author.  I like his 'mfwd' instruction, I have
every intention of incorporating that idea.  Till now it's been a relatively
low priority.  If your willing to work with me I might be willing to
move it up the priority list :-)

> =20
> Worth trying ?? ;o)
> =20
> Cheers,
> Jonathan
>=20
> 	-----Original Message-----=20
> 	From: James R. Leu [mailto:jl...@mi...]=20
> 	Sent: Fri 3/09/2004 11:44 PM=20
> 	To: Chan, Jonathan (ICT Centre, Marsfield)=20
> 	Cc: mpl...@li...=20
> 	Subject: Re: [mpls-linux-general] Multicast routing over linux mpls tunn=
el ?
> =09
> =09
>=20
> 	I haven't dealt with multicast much, but I do know it is real picky
> 	about what interface a packet appears to arrive on.  With MPLS tunnels,
> 	packets you receive off an LSP appear to come from the physical interface
> 	(eth0 etc)  I think pimd is expecting to see the packets come from the s=
ame
> 	interface it sent them out (mpls0 etc).  I think I added code at one poi=
nt
> 	to work around this.
> =09
> 	mplsadm2 -L eth0:0
> 	mplsadm2 -A -I gen:16:0
> 	mplsadm2 -I gen:16:0 -i set:mpls0:pop:peek
> =09
> 	I just looked at the code for mplsadm2 and it won't work without
> 	some modifications, I'm attaching a patch against the latest version (1.=
935)
> 	you can get the source for mplsadm2 at http://sf.net/projects/mpls-linux
> =09
> 	Apply the patch by going into the mpls-linux/utils directory and execute
> 	patch mplsadm.c < mplsadm.patch
> =09
> 	then issue a 'make'
> =09
> 	(make sure your system header files include mpls.h  if not get
> 	my glibc-headers RPM from http://sf.net/projects/mpls-linux)
> =09
> =09
> 	On Fri, Sep 03, 2004 at 05:10:26PM +1000, Jon...@cs... wrote:
> 	> Dear all,
> 	>
> 	> I don't know if I am asking for something impossible... Just wondering
> 	> if someone out there might have experience/suggestion to get me to the
> 	> right ball park...
> 	>
> 	> I have two linux mpls boxes linked by two mpls tunnels (one at each
> 	> direction, of course ;o). I managed to install pimd to handle multicast
> 	> routing. Is there anyway to make the linux mpls tunnel carry the igmp
> 	> traffic??
> 	>
> 	> Currently, my linux mpls was created by:
> 	>
> 	> % mplsadm2 -A -O 0
> 	> % mplsadm2 -O $opcode -o push:gen:123:set:eth0:ipv4:192.168.2.1
> 	> % mplsadm2 -A -T tun21
> 	> % ifconfig tun21 192.168.1.1 netmask 255.255.255.255 pointopoint
> 	> 192.168.2.1
> 	>
> 	> I also made this tunnel multicast enabled.
> 	>
> 	> % ip link set tun21 multicast on
> 	>
> 	> But I don't know what I need to do to enable the tunnel to carry igmp
> 	> traffic :o(
> 	>
> 	> Do I have to put in GRE tunnel or something??
> 	>
> 	> Thanks for your time and consideration...
> 	>
> 	> Cheers,
> 	> Jonathan
> 	>
> 	>
> 	>
> 	> -------------------------------------------------------
> 	> This SF.Net email is sponsored by BEA Weblogic Workshop
> 	> FREE Java Enterprise J2EE developer tools!
> 	> Get your free copy of BEA WebLogic Workshop 8.1 today.
> 	> http://ads.osdn.com/?ad_idP47&alloc_id=10808&op?k
> 	> _______________________________________________
> 	> mpls-linux-general mailing list
> 	> mpl...@li...
> 	> https://lists.sourceforge.net/lists/listinfo/mpls-linux-general
> =09
> 	--
> 	James R. Leu
> 	jl...@mi...
> =09
>=20
> ???????????????????????????????????????????+=12=17?^????X???'???u??=10=05=
?nZ ????K!??Q=10BZ??'????+=1E'a=04u????^??(???z?????y?(?*=1F=04@=16y???'=16=
??,??|?[hu????i???v?????r?????????jYhs?????????????????????????????????????=
???????????{??=07?z??????x%??f?[??)???=1E?????b??????q???=07???a??l?????l??=
.?????=1E?w??X??????b?????l?X??=1F?zw?

--=20
James R. Leu
jl...@mi...