[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: IGPs and LMP's TE-link
try to clarify the question with an example:
have 4 data-links a,b,c,d.
define 2 TE-links: TEL1 containing a&b, TEL2 containing
components may the UNI use? for
components may the IGP use? for
components may the TE engine use
for CSPF? for which purposes?
'logical' interfaces will the local
UNI-C have towards the remote UNI-C? a,b,c &d or TEL1 &
I've found your questions rather confusing. Any way, the following are the
Hope this helps.
- There is no difference (as such) between the way a TE link is
represented by LMP and within IGP (with the exception of the level of
details each protocol is concerned with).
- There is no direct relation between UNI 1.0 and IGP. However, once UNI
connection(s) is (are) established, one may run IGPs on that (these)
connections (either in a bundled or unbundled configuration). But this case
does not require any standard specification.
- Multiple component links between UNI-C and UNI-N can be regarded as TE
links without an IGP adjacency. But one can run LMP on such TE Links.
At 08:06 PM 5/22/2002 +0300, Liran Siglat wrote:
I have a question regarding the way
LMP's TE-link is observed by IGPs.
In case a connection was
established between 2 UNI-Cs over a TE-link (aggregating multiple
data-bearing links), does an IGP running on the UNI-C see the TE-link as a
new interface (with the capacity of all the data-links) or does it see
multiple interfaces (each one corresponds to a data-link)
100 S Main St. #200
Ann Arbor, MI
Mobile: (734) 276-2459, Off: (734) 302-4143, Fax: (734)