From: tom cheung (tkc9789@xxxxxxxxxxx)
Date: Sat Nov 17 2001 - 23:30:45 GMT-3
I'll have to disagree, and someone else please correct me if I'm wrong.
OSPF only understands intra-area, inter-area and external route types. OSPF
route preference is based upon route types and within the same route type,
costs. If there're equal cost routes within the same route type, then load
balance is automatic. Sure you can look at same cost as hop counts. If I
make the cost of two paths to the same destination, say 100, you certainly
can view that as 100 hops. But load balance will happen.
If there're no equal cost paths to the same destination, fast switch will
not put packets on the higher cost path, period.
Tom
>From: Hansang Bae <hbae@nyc.rr.com>
>Reply-To: Hansang Bae <hbae@nyc.rr.com>
>To: ccielab@groupstudy.com
>Subject: Re: OSPF Load Balance
>Date: Sat, 17 Nov 2001 20:14:20 -0500
>
>>>From: Hansang Bae <hbae@nyc.rr.com>
>>>I don't think you accomplished what you wanted to do. Setting all the
>>>costs to be the same causes OSPF to act like RIP. That is, the metric
>>>now
>>>becomes a hop count. So unless you have the same number of hops, it
>>>won't
>>>work.
>>>Even then, route-cache will cause per destination load-balancing.
>
>
>At 10:17 AM 11/17/2001 -0600, tom cheung wrote:
>>Correct me if I'm wrong. By default, OSPF load balances up to a max of 4
>>equal cost, equal path-type routes. OSPF does not take into consideration
>>of hop counts.
>
>
>
>All correct. *Except*, that the original poster made all the interface
>costs to be the same. So basically, the metric acts just like hop counts.
>And while load balancing works across 4 links (w/o using maximum-paths),
>the route-cache (fast processing) makes the router use one link for a
>particular destination. This isn't unique to OSPF, it's just the way fast
>processing works (w/o using CEF etc.)
>
>hsb
This archive was generated by hypermail 2.1.4 : Fri Jun 21 2002 - 06:45:17 GMT-3