On Thu, Mar 29, 2007 at 02:33:55PM -0500, [EMAIL PROTECTED] wrote:
> Hi Rodney,
> 
> Do you have details of problems related to process scheduling in 76xx
> when using aggressive RSVP hellos?

Basically if some other process is running there is a chance you can't
get to the rsp hello message processing fast enough so you would declare
the link down when really it's not.

> 
> The pseudo-preemption you mentioned, could you give more details?

Not really. Just a way for the scheduler to signal that a process needs
to run now versus just being scheduled like another process.

> 
> I agree that monitoring loss of signal is the preferred way to do it;
> but for cases where there is no LOS capability, do we have other option
> besides RSVP hellos?

No. You can use them you just can't tune them too low or you run the risk
of getting false positives.

> (considering BFD for FRR is not ready to use) 
> 
> Tks,
> Alaerte  
> 
> -----Original Message-----
> From: ext Rodney Dunn [mailto:[EMAIL PROTECTED] 
> Sent: Thursday, March 29, 2007 4:26 PM
> To: Vidali Alaerte (Nokia-NET/RioDeJaneiro)
> Cc: [email protected]
> Subject: Re: [c-nsp] RSVP Hellos
> 
> None other than the standard issues of subsecond hellos and process
> scheduling. We implemented a pseudo-preemption type mechanism for BFD to
> get fast reaction.
> 
> RSVP fast hellos isn't the preferred way to do it.
> 
> LOS is.
> 
> There was/is some talk around BFD triggered FRR.
> 
> No idea what that roadmap is though.
> 
> Rodney
> 
> On Thu, Mar 29, 2007 at 02:07:19PM -0500, [EMAIL PROTECTED]
> wrote:
> >  
> > Are you aware of any restriction concerning use of aggressive values 
> > on RSVP Hellos to detect neighbor failure?
> > 
> > 
> > 
> > 
> > _______________________________________________
> > cisco-nsp mailing list  [email protected] 
> > https://puck.nether.net/mailman/listinfo/cisco-nsp
> > archive at http://puck.nether.net/pipermail/cisco-nsp/
_______________________________________________
cisco-nsp mailing list  [email protected]
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/

Reply via email to