> Sorry, I’m not following you. If a link is not power sleep capable, > then it cannot be put to sleep or the network operator does not want it put to sleep.
Obvious. But the way I read what you said was different. I assumed that 100% of links must advertise power related information in order for them to be eligible and become a player in an TE computation game. I hope this is not the case. Thx, R. On Sat, Aug 29, 2026 at 7:16 PM Tony Li <[email protected]> wrote: > Hi Robert, > > > > If an interface is not power sleep capable, then its PSP is not > relevant in metric computation. > > > > Why would you say so ? > > > That’s the semantics that we’re proposing. > > > > I can see some links (or even entire parts of topologies) > > which do not have any power saving potential configured but they could > be used just fine for taking some traffic from links designated for > sleeping under lower load. > > > Sorry, I’m not following you. If a link is not power sleep capable, then > it cannot be put to sleep or the network operator does not want it put to > sleep. The PSP is not used in the path metric computation, so the link > becomes a (relative) traffic attractor, drawing traffic away from power > sleep capable links. > > > > If you assume that only links participating in power savings should be > part of it I think this is significantly problematic. > > > Operators are very clear: they need a minimum resilient topology that > ensures that they have a redundant network at all times. There are links > that must not be put to sleep. This seems quite sensible. > > Tony > > >
_______________________________________________ Lsr mailing list -- [email protected] To unsubscribe send an email to [email protected]
