[j-nsp] Ethernet OAM question?

[Humair Ali]

Hi Tim

not sure what you mean by using LACP/AE accross the multiple network to R2 ,
as LACP PDU does not go beyond the first physical interface it encounters,
unless as mention by David, you use some kind of tunneling, but then again I
am not sure how it works accross providers.

But David is right , CFM is the way to go in your setup,

To bring the interface down, you should be able to create an "action profile
" under CFM , I am sure there is an option to bring the link down and you
can then apply the action profile

Cheers

On 19 October 2011 06:30, David Ball <davidtball at gmail.com> wrote:

> On 18 October 2011 22:46, tim tiriche <tim.tiriche at gmail.com> wrote:
> > Can i run link fault management between R1-R2 or is LFM only between
> > R1 - switchA?
>
>   The latter, as LFM PDUs shouldn't be forwarded by switches.  It
> should be link-local.
>
> > What keepalives method i can use to detect a failure somewhere in the
> > provider and bring the R1-R2 physical link down.
> > Bring this link as AE and run LACP?
>
>   OAM CFM (connectivity fault management) could be used for this, if
> R1 and R2 support it, which is designed to provide end-to-end OAM
> monitoring via periodic CCMs (continuity check messages).  We're
> seeing better support for this in the 11.2 (or was is .3?) or later
> code on MXs.  You may need an event script to actually tear your link
> down locally if that's what you're looking for (based on a syslog
> event, for example), as OAM CFM doesn't do that for you.
>
>  I wouldn't trust the intermediary providers to transport
> 802.3ad/LACP PDUs transparently, unless they all have indicated that
> it's supported via L2 protocol tunneling (though I'd still be a
> skeptic, if there are multiple carriers).
>
> David
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/juniper-nsp
>



-- 
Humair