Re: [OSPF] OSPF NSSA ABR behaviors

["Pat Murphy - [phone]" <[e-mail]>]

James,

See comments inline.

P>Thus if one desires to aggregate the external route during Type-7
P>translation using a Type-7 address range then a Type-7 LSA with the P-bit
P>set must be originated.

P>Note the latter statement from RFC 3101 simply ensures that non-NSSA
P>traffic forwarded by a Type-5 LSA installation toward a forwarding address
P>inside an NSSA won't loop. I prefer originating the Type-7 LSA with the
P>P-bit set and simply letting the Type-7 translation process originate the
P>Type-5 LSA. Thus Type-5 LSAs would then never need to set a forwarding
P>address that belongs in a stub area or an NSSA, except during the Type-7
P>LSA translation process. Hence if the best forwarding address is in the
P>transit topology, it is better to originate both a Type-5 LSA and a Type-7
P>LSA with the P-bit clear. If the best forwarding address belongs to the
P>NSSA, it is better to simply originate a Type-7 LSA with the P-bit set and
P>let the Type-7 translation process originate the Type-5 LSA.

J>When you say a router originates a type-7 LSA, does it mean
J>that the type-7 LSA will be advertised?

Of course, all LSAs are advertised by an originating router.
See RFC 2328 Appendix A.4.1, "The LSA header".

J>Or it simply means that the type-7 LSA is generated internally
J>and it depends on the translation process to determine whether
J>it should be aggregated with some other external LSA(s) or be
J>advertised individually? 

Like all LSAs, Type-7 LSAs are only generated to be originated
by an advertising router. But once originated with the P-bit set
the Type-7 LSA may generate a Type-5 translation as described in
RFC 3101 Section 3.2, Translating Type-7 LSAs into Type-5 LSAs.

P>Let's assume that the redistribute is intended to feed the route
P>into all NSSAs and the non-NSSA topology. If (G) belongs to the non-NSSA
P>topology, originate a Type-5 LSA with forwarding address (G) into the
P>non-NSSA topology and also originate a Type-7 LSA with the P-bit clear
P>into each NSSA. Here the Type-7 LSAs can simply use a 0.0.0.0 forwarding
P>address. If (G) belongs to one of the NSSAs, then originate a Type-7 LSA
P>with the P-bit set into that NSSA with the forwarding address set to (G).
P>Finally originate a Type-7 LSA with the P-bit clear into the remaining
P>NSSAs. Note again that Type-7 translation will cause the appropriate
P>Type-5 LSA to be originated into the non-NSSA topology.

J>Agree - assuming address aggregation is not performed.

Type-7 ranges that subsume only a single prefix with the same
[address,mask] pair result in a normal translation. Hence, since
type-7 ranges can even subsume other type-7 ranges, one can negate
the effects of aggregation under configuration control. This is a
subtle point from RFC 3101 Section 3.2.

J>On a related question: When a MSSA ABR originates a aggregated
J>type-5 LSA, does it has to advertise a corresponding type-7 (p=0)
J>LSA into the NSSA area?  I cannot find any rule in RFC 3101 about this. 

The origination of a type-5 LSA generated by the translation of a
type-7 LSA NEVER forces the origination of another type-7 LSA.

J>I think the answer is no unless more than one locally sourced
J>external LSAs with the best forwarding addresses NOT in the NSSA are
J>aggregated into the type-5 LSA (otherwise, individual type-7 LSA
J>(p=0) should be advertised).   Is this correct?

The non-zero forwarding addresses of Type-7 LSAs are always
chosen from interfaces that belong to the NSSA. See Section 2.3.
A type-5 translation that aggregates a set of installed type-7
LSAs also NEVER forces the origination of another type-7 LSA.

J>One more question, RFC3101, sec 3.2 step (3):

          The path type and metric of the Type-7 address range is
          determined from the path types and metrics of those
          translatable Type-7 LSAs that best match the range plus any
          locally sourced Type-5 LSAs whose network has the same
          [address,mask] pair.

J>What is the reason aggregation can only aggregate those locally sourced
J>type-5 LSAs with the same [address, mask] as the configured address range?

Since this is an aggregation its metric must reflect the highest cost 
amongst the type-7 LSAs being aggregated. A locally source type-5 LSA
that has the same [address,mask] pair coupled with a lower cost could
negate the translation metrics. Similarly a lower cost in the Type-5
translation of a type-7 range would negate the high cost of a
locally sourced type-5 LSA with the same [address,mask] pair. Hence
a single advertisement is required.

J>If a locally sourced type-5 LSA is subsumed within the range, can it
J>be aggregated too?

No. The above excerpt is the only exception.

Pat