From: Martinez, Paul (PMartinez@xxxxxxxxxxx)
Date: Thu Mar 30 2000 - 14:14:29 GMT-3
Title: RE: OSPF costs
There were no Gig interfaces or FDDI for that mater,
when I took the lab.. however the equipment in the rack
does support it.
You bring up an interesting point though,
i would say that altering the cost of a FDDI/FastE link would
be the only solution although it doesn't scale very well in
a large enterprise. Generally Gig interfaces are not as widelly
deployed as Fast/E (although give it a few years and...) one could
alter the ref-bandwidth OSPF uses to calculate the cost so that
all FDDI/FastE interfaces are greater than 1 then alter the cost
of each Gig interface to be less than FDDI/FastE intefaces.
Anyone else?
-----Original Message-----
From: ejobson
To: ccielab@groupstudy.com
Sent: 3/30/00 11:25 AM
Subject: OSPF costs
I wonder if someone could help me out with this:
I understand that the costs are calculated for OSPF by
100,000,000/bandwidth
which gives a standard ethernet connection a value of 10. A fast
ethernet or
FDDI would be 1, so I guess a Gigabit link would also be 1. What
attribute
would be used by default to ensure traffic took the 1000Mbps over the
100Mbps one, or is it just a limitation due to OSPF now being a bit
dated? I
have heard a solution for this in the past but have since forgotten
it.
Is it a case of manually setting the cost of all fast ethernet
interfaces in
the network to say 5, as it doesn't seem possible to have a value
higher
than 1.
Also is there any chance of Gigabit being used to confuse OSPF
calculations
in the exam?
Thanks.
Eddie Jobson
Internetwork Solutions Engineer
Nevolutions UK Ltd. a division of
Total Network Solutions Inc.
7th Floor, Tower 42
The International Finance Centre
25 Old Broad Street
London
EC2N 1HN
Office +44 (0)171 877 1992
Fax +44 (0)171 877 1059
Mobile +44 (0)7867 505635
Email ejobson@tns-inc.com
Website http://www.tns-inc.com
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:23:07 GMT-3