[j-nsp] RSVP LSP Bandwidth
Chris Grundemann
cgrundemann at gmail.com
Mon Jan 19 11:59:13 EST 2009
On Mon, Jan 19, 2009 at 09:06, Andrew Jimmy <good1 at live.com> wrote:
>
> What does meant by reserved and highwater bandwidth reservation exactly. Is
> this something similar to police rate-limit or traffic shaping? Why do you
> need to reserve bandwidth and in what circumstances ; what are the impacts?
>
>
>
> I will appreciate if someone can explain all of this in very simple words.
>
>
>
> juniper at nasa> show rsvp interface
>
> RSVP interface: 2 active
>
> Active Subscr- Static Available
> Reserved Highwater
>
> Interface State resv iption BW BW BW
> mark
>
> so-0/0/0.0 Up 0 100% 155.52Mbps 155.52Mbps 0bps
> 0bps
>
> so-0/0/2.0 Up 1 100% 155.52Mbps 105.52Mbps 50Mbps
> 50Mbps.
Reserved Bandwidth is the amount of bw currently requested/reserved by
all LSPs using that interface. Highwater Bandwidth is the highest
level that Reserved Bandwidth has reached (since the statistics were
last cleared). A full explanation of the output fields can be found
here: https://www.juniper.net/techpubs/software/junos/junos91/swcmdref-protocols/show-rsvp-interface.html
(don't worry about the code version - this output has not changed much
over a lot of versions).
RSVP is the resource reservation protocol; when used for MPLS the
resource it reserves is interface bandwidth (and a label). This is
not for shaping or policing; this is part of how an LSPs path is
determined. The bandwidth reservation is used for traffic
engineering, in its simplest form this keeps new LSPs from being
signaled across (potentially) saturated links.
I highly recommend that you read chapters 7 and 8 of the JNCIS-M study
guide (pdf available free:
http://www.juniper.net/training/certification/JNCIS_studyguide.pdf)
for more information on MPLS, RSVP and Traffic Engineering (TE) in
JunOS.
>
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/juniper-nsp
--
Chris Grundemann
www.chrisgrundemann.com
More information about the juniper-nsp
mailing list