Re: a doubt on SPF

[Acee Lindem <[e-mail]>]

Alok,

See below - I attempted to answer the two questions
at the end.

Alok Dube wrote:
> Hi Acee,
>
> thanks for your reply...
>
> maybe im a bit grey but ths thing is stil fuzzing me up...
>
> inline...
>
>
>>Alok Dube wrote:
>>
>>>Hi,
>>
>>Hi Alok,
>>
>>
>>>Assume I have 2 ABRs...
>>>
>>>Area 1  connects to Area 0 via ABR-1 and ABR-2.
>>>
>>>my questions are as follows.
>>>
>>>1. will ABR-2 learn the route to Area 0 via ABR-1 (network LSA)...i
>>>think it should..correct?
>>
>>ABR-2 should use it's intra-area path to routes in area 0. This
>>question is a bit ambiguous due to the fact that there is no network
>>diagram.
>
>
> i dont have a diagram on hand, its basically a question as to why should
> "area boundarys" define the "shortest path"....
>
>>
>>>2. if the path to a destination is better via ABR-2----(router in Area
>>>1)--ABR-1---area 0 routers---(ABR of other Area), it will take this
>>>route..compared to ABR-2 -- area 0 routers ---(ABR of other Area)...am
>>>i correct?
>>
>>OSPF Routers will prefer intra-area paths over inter-area paths.
>
>
>
> even if the inter area has a better SPF calculated metric from
> ABR-2?..thats strange...in a way, isnt it?
>
>
>>
>>>3. if the destination was in Area 0 and the above condition again held
>>>true, there is no rule that a route learnt via type 1 lsa is given
>>>precedence over type 3?
>>
>>
>>See #2.
>>
>>
>>>3. Does OSPF ensure that routes learnt from a router are nevr sent
>>>back (split horizon types)...can someone point me to the  part of RFC
>>>2328 that states the same? or is the method implemented by means of
>>>simply checking the originator RID and discarding the LSA if it
>>>belongs to oneself?
>>
>>This isn't needed for intra-area routes - the SPF calculation will
>>ensure there is no looping. For generated summary advertisements, one
>>doesn't advertise them back into the area corresponding to the route.
>
>
>
> fine, i saw the part on summary LSAs, and the fact that OSPF wont pass on
> routes which belong to an area back into the same area via another ABR...
> but on an ABR, would i use it?....
>
> for example on the ABR-2, I have 2 interfaces, int-1 in Area 1 and int-2
> in Area 0.
> Now if the router is a "datacentre" switch too, and is a next-hop for
> various servers to get to the other networks, shouldnt it use the better
> SPF path via ABR-1 to get to a destination...?
>
>
>>
>>>4. if there are 2 ABRs, how will the LSAs originated by ABR-1 be
>>>treated by ABR-2? which section of the RFC deals with this?..is page
>>>135 line 1 the answer?..but does the ABR itself use this LSA for its
>>>SPF
>>>calculations?(in a way this is same as case 1 above)
>>
>>Read section 16.2 in RFC 2328. All the cases are covered.
>>
>
>
>
> 16.2 point 3
> 3) If it is a Type 3 summary-LSA, and the collection of
>           destinations described by the summary-LSA equals one of the
>           router's configured area address ranges (see Section 3.5),
>           and the particular area address range is active, then the
>           summary-LSA should be ignored.  "Active" means that there
>           are one or more reachable (by intra-area paths) networks
>           contained in the area range.
>
>
> which means ignore it if its a route to area-0...
>
> if its a route to another area which is learnt via both area-0 and ABR-1
> (area 1)........... then what?
> would :
> 4) Else, call the destination described by the        LSA N (for Type
>           3 summary-LSAs, N's address is obtained by masking the LSA's
>           Link State ID with the network/subnet mask contained in the
>           body of the LSA), and the area border originating the LSA
>           BR.  Look up the routing table entry for BR having Area A as
>           its associated area.  If no such entry exists for router BR
>           (i.e., BR is unreachable in Area A), do nothing with this
>           LSA and consider the next in the list.  Else, this LSA
>           describes an inter-area path to destination N, whose cost is
>           the distance to BR plus the cost specified in the LSA. Call
>           the cost of this inter-area path IAC.
>
> will this come in here?
>
> or would it be as in 16.3 :
>
> 5) If this cost is less than the cost occurring in N's        routing
>           table entry, overwrite N's list of next hops with those used
>           for BR, and set N's routing table cost to IAC. Else, if IAC
>           is the same as N's current cost, add BR's list of next hops
>           to N's list of next hops. In any case, the area associated
>           with N's routing table entry must remain the backbone area,
>           and the path type (either intra-area or inter-area) must
>           also remain the same.
>
> ....isnt this a rule which primarily restricts OSPF from having "area in
> an area"?
>
> there are 2 things im trying to clarify
>
> 1. what tells SPF that "use the backbone route"

When you do your summary calculation, routers with multiple areas only
consider backbone summaries (section 16.2). draft-ietf-ospf-abr-alt-05.txt
allows summaries from multiple areas to be considered in situations where the
router is detached from the backbone area.

> 2. why can cant i simply have a "ring of areas" and no area 0? (bandwidth
> constraints/LSA flooding setaside for my drawing board model...).

The protocol as described in RFC 2328 affords deterministic and loop free
operation between interoperating OSPF routers given the documented
SPF route calculation (as described in section 16). The "ring of areas"
topology is not supported by either RFC 2328 or draft-ietf-ospf-abr-alt-05.txt.
If you want to design an OSPF network with multiple area your design MUST
include a backbone area.

Thanks,
Acee

> -rgds
> Alok
>
>
>
>
>
>
>
>
>


--
Acee