Standard Track.
Referring to TE LSP signalled.
When you signal a TE LSP with zero bandwidth then the TLVs used to advertise the reserved bandwidth are obviously unchanged.
Thanks.
JP.
thanks,
- dimitri
05/09/2006 00:00
Please respond to Adrian Farrel
To:
cc:
Subject: Re: [mpls] WG Last Call on
draft-ietf-mpls-number-0-bw-te-lsps-02.txt
Hi,
A quick review of this I-D leaves me puzzled by the motivation. I don't
think it is enough to say "I want to know how many unconstrained TE LSPs
traverse every link in the network"; we must have an idea of how that
information will be used.
Now the I-D says:
... in the Abstract
There are various
circumstances (for example in order to load balance unconstrained TE
Label Switched Path (LSP) across a set of equal cost paths) where it
would be useful to also advertise the number of unconstrained Traffic
Engineering Label Switched Path(s) (TE LSP) signalled across a link.
... and in the Introduction
If the
number of unconstrained TE LSPs traversing each link in the network
is known, various algorithms can be designed so as to efficiently
load balance the traffic carried onto such unconstrained TE LSPs.
...but this seems to me to be an supported assertion. I guess you could
cite
a reference so you don't need to prove the algorithms here. But it seems
to
me that the count of such LSPs is a very poor measure indeed. The only way
you could use it would be if you could make some fairly tight statistical
assumptions about the traffic on each LSP, or at least the statistically
aggregate traffic on a number of LSPs.
Surely it would be better for the LSRs to report link usage rather than
the
number of LSPs? If you are trying to protect the LFIB from excessive
growth
then your proposal would be reasonable, but the problem you have stated is
one of load balancing and for that you need to know the load, not the
number
of LSPs.
I also have the following nits and small issues in the I-D.
A Link-Type sub-TLV to convey the number of Traffic Engineering Label
Switch Paths signalled across a link
## Is this title right? I think your proposal is only to count LSPs that
have zero reserved bandwidth.
Abstract
## I think "ISIS" should be spelled "IS-IS"
## Some acronyms need to be expanded on their first usage:
## IS-IS, MPLS, TE
1. Introduction
## I think "ISIS" should be spelled "IS-IS"
## Acronyms have to be spelled out again (separate from the Abstract) the
## first time they are used. Maybe move the terminology section above the
## Introduction?
A set of Link-type sub-TLVs have been defined for OSPF and ISIS (see
[I-D.ietf-isis-te-bis] and [RFC3630]) in the context of MPLS Traffic
## References are in the wrong order
It is not uncommon to deploy MPLS Traffic Engineering for the sake of
fast recovery relying on a local protection recovery mechanism such
as MPLS TE Fast Reroute (see [RFC4090]). In this case, a deployment
model consists of deploying a full mesh of unconstrained TE LSPs
between a set of LSRs and protect these TE LSPs with pre-established
## s/protct/protecting/
## Better to reorder..
## ...and protecting these TE LSPs against link..... with
pre-established...
backup tunnels against link, SRLG and/or node failures. The traffic
routed onto such unconstrained TE LSP simply follows the IGP shortest
path but is protected with MPLS TE Fast Reroute.
## I guess you are implying that each TE LSP is installed as a virtual
link
## in the IGP and is assigned a cost. But you haven't said so.
With MPLS Traffic Engineering a usual rerouting criteria is the
discovery of a better path for a TE LSP where a better path is
defined as a path with a lower cost according to a specific metric;
other metric such that the percentage of reserved bandwidth or the
number of hops can also be used.
## I can guess the meaning here, but I can't parse the sentence.
Unfortunately, for instance in the
presence of ECMPs (Equal Cost Multi-Paths) in symmetrical networks
when unconstrained TE LSP are used, such metrics are usually
## s/LSP/LSPs/
This document specifies a new Link-type Traffic
Engineering sub-TLV used to indicate the number of unconstrained TE
LSP signalled across a link.
## s/LSP/LSPs/
Note that the specification of load balancing algorithms is outside
of the scope of this document and merely listed for the sake of
illustration of the motivation for gathering such information.
## Very good, but if such algorithms cannot exist based on the data you
## are collecting, then you would do better to list another illustrative
## example.
Furthermore, the knowledge of the number of unconstrained TE LSPs
signalled across each link can be used for other purposes (e.g.
management, ...).
## Either make some good suggestions or delete this paragraph. But
## that would be to note that the only use you have to suggest is
## "load balancing"
2. Terminology
Unconstrained TE LSP: A TE LSP signalled with a bandwidth equal to 0.
## When you say "a bandwidth equal to 0" you do not mean that there is
## no traffic carried. Can you be more specific in this definition about
## what you do mean.
3.1. IS-IS
The NB-0-BW-LSP sub-TLV is OPTIONAL and MUST appear at most once
within the extended IS reachability TLV (type 22) specified in
[I-D.ietf-isis-te-bis].
## What should a receiver do if a second instance is encountered?
Value (4 octets): number of unconstrained TE LSP(s) signalled across
the link.
## Do you intend to only count the TE LSPs that have been signaled or
## are you also interested in those that have been manually configured?
3.2. OSPF
The NB-0-BW-LSP sub-TLV is OPTIONAL and MUST appear at most once
within the Link TLV (Type 2) that is itself carried within the
Traffic Engineering LSA specified in [RFC3630].
## As above
Value (4 octets): number of unconstrained TE LSP(s) signalled across
the link.
## As above
4. Elements of procedure
An implementation may decide to implement a dual-thresholds mechanism
to govern the origination of updated OSPF LSA or ISIS LSP. Similarly
to other MPLS Traffic Engineering link characteristics, LSA/LSP
origination trigger mechanisms are outside of the scope of this
document.
## Is that "may" actually a "MAY"? Wouldn't a SHOULD or MUST be better?
## If the unconstrained TE LSPs are stable (i.e. not changing often) then
## there is clearly no need for this extension as part of the routing
## protocol - it could be safely gathered using the management plane. So
## we must assume that the rate of change of unconstrained TE LSPs is
## non-trivial in which case this I-D must concern itself with the impact
## on the stability of the IGP of distributing this information.
5. IANA Considerations
## Can you point IANA at the appropriate registries, please.
6. Security Considerations
This document raises no new security issues for IS-IS and OSPF.
## You are distributing additional information about the network, so
## there must be an additional security consideration.
## What would happen if an LSR accientally or deliberately misreported
## the number of LSPs? What wuld happen if that information was spoofed
## or interfered with?
8.1. Normative References
[I-D.ietf-isis-te-bis]
Li, T. and H. Smit, "IS-IS extensions for Traffic
Engineering", draft-ietf-isis-te-bis-00 (work in
progress), September 2005.
## Is this date right? I see August 2005.
## Anyway, the I-D has expired which is not a good sign for a normative
reference.
Cheers,
Adrian
----- Original Message -----
Sent: Friday, September 01, 2006 9:08 AM
Subject: [mpls] WG Last Call on draft-ietf-mpls-number-0-bw-te-lsps-02.txt
Working Group,
this initiates a two week working group last call on
draft-ietf-mpls-number-0-bw-te-lsps-02.txt
The wg last call ends on September 17.
Please send comments to the working group mailing list and/or
the working group chairs.
/Loa and George
--
Loa Andersson
Principal Networking Architect
Acreo AB phone: +46 8 632 77 14
Isafjordsgatan 22 mobile: +46 739 81 21 64
_______________________________________________
mpls mailing list
_______________________________________________
mpls mailing list
_______________________________________________
mpls mailing list