From: Paul Cosgrove (paul.cosgrove@heanet.ie)
Date: Sat Dec 29 2007 - 06:58:51 ART
The behaviour regarding not not prefering locally originated LSAs is
specified in RFC2328:
16.4. Calculating AS external routes
AS external routes are calculated by examining AS-external-LSAs.
Each of the AS-external-LSAs is considered in turn. Most AS-
external-LSAs describe routes to specific IP destinations. An
AS-external-LSA can also describe a default route for the
Autonomous System (Destination ID = DefaultDestination,
network/subnet mask = 0x00000000). For each AS-external-LSA:
(1) If the cost specified by the LSA is LSInfinity, or if the
LSA's LS age is equal to MaxAge, then examine the next LSA.
(2) If the LSA was originated by the calculating router itself,
examine the next LSA.
The last point I mentioned turns out to be incorrect; just reading the
rfc's and type-7 LSAs always have the forwarding address set.
Appologies for resurrecting this month old thread, should have checked
the date befre replying!
Paul.
Paul Cosgrove wrote:
> The quote seems to refer to removal of routes, rather than the
> advertisement of them. It seems to be referring to a situation when
> you cease receiving one of possibly several type-7 LSAs.
> In Josephs case the translating ABR still has both the type-5 and the
> type-7 in its OSPF database so this is a little different.
> The reason for the behaviour is that R1 will not prefer a type-5 LSA
> it itself has created locally, over the type-7 it used to create it.
> If the forwarding address is not set then to do so would cause R1 to
> send the traffic to itself, rather than to the ASBR (i.e. create a
> blackhole). As mentioned only the ABR with the highest IP creates a
> type-5, so this ABR will not receive a similar type-5 LSA from elsewhere.
>
> Paul.
>>
>> On Nov 11, 2007 12:15 AM, shiran guez <shiranp3@gmail.com> wrote:
>>
>>> and here is why you see only on one of the ABR's the N2:
>>>
>>> 3.3 Flushing Translated Type-7 LSAs
>>>
>>> If an NSSA border router has either translated or aggregated an
>>> installed Type-7 LSA into a Type-5 LSA that should no longer be
>>> translated or aggregated, then the Type-5 LSA should either be
>>> flushed or reoriginated as a translation or aggregation of other
>>> Type-7 LSAs.
>>>
>>>
>>>
>>>
>>> On Nov 10, 2007 10:14 PM, shiran guez <shiranp3@gmail.com> wrote:
>>>
>>>
>>>> Here is a quate from RFC 3101, this is what i mean by saying that the
>>>>
>>> concept is only one exit for stub and NSSA.
>>>
>>>> 3.1 Type-7 Translator Election
>>>>
>>>> It is not recommended that multiple NSSA border routers perform
>>>> Type-7 to Type-5 translation unless it is required to route packets
>>>> efficiently through Area 0 to an NSSA partitioned by Type-7 address
>>>> ranges. It is normally sufficient to have only one NSSA border
>>>> router perform the translation.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On Nov 10, 2007 10:10 PM, Joseph Saad <joseph.samir.saad@gmail.com>
>>>> wrote:
>>>>
>>>>
>>>>> I just disagree with the statement:
>>>>>
>>>>>
>>>>> the concept of NSSA and Stub Areas is one exit only.
>>>>>
>>>>> The concept of NSSA is to eliminate type 3, 4 and 5 LSA's while
>>>>> allowing redistribution via having type 7.
>>>>> The concept of stub area is to eliminate type 3,4,5 and never allow
>>>>> redistribution.
>>>>>
>>>>> There's no such mandate on the number of ABR (exits) to be one.
>>>>>
>>>>> Anyhow, reading RFC 3101 for enlightenment.
>>>>>
>>>>> cheers,
>>>>> Joseph.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Nov 11, 2007 12:04 AM, shiran guez <shiranp3@gmail.com> wrote:
>>>>>
>>>>>> R1 is the ABR for the NSSA so you will see the N2 sent from R6
>>>>>> ASBR as
>>>>>>
>>> you
>>>
>>>>>> see now, and R1 as you can see in the DB did the translation for
>>>>>> area
>>>>>>
>>> 0 and
>>>
>>>>>> inside Area 0 on R5 you should see the E2, there is no preferance of
>>>>>>
>>> N2 over
>>>
>>>>>> E2.
>>>>>>
>>>>>> N2 is exactly the same as E2 but for routes that came from NSSA.
>>>>>>
>>>>>> R1 is the translator meaning that he recive from R6 and you see the
>>>>>>
>>> routes
>>>
>>>>>> recived on the routing table as they are valid and applying to the
>>>>>>
>>> linkstate
>>>
>>>>>> rules. and then R1 is translating and sending out to area 0.
>>>>>>
>>>>>> the concept of NSSA and Stub Areas is one exit only.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Nov 10, 2007 9:32 PM, Joseph Saad <joseph.samir.saad@gmail.com >
>>>>>>
>>> wrote:
>>>
>>>>>>>
>>>>>>> Here's an elaboration on the issue.
>>>>>>>
>>>>>>> R1
>>>>>>>
>>>>>>> R6 (A1) (A0) R5.
>>>>>>>
>>>>>>> R4
>>>>>>>
>>>>>>> R1, R4 and R6 shares the same VLAN and is in area 1.
>>>>>>> R1, R4 and R5 are on a Hub-n-spoke with R5 as the hub and is in
>>>>>>> area
>>>>>>>
>>> 0
>>>
>>>>>>> R6 is the ASBR (where I am doing redistribution of static subnets).
>>>>>>>
>>>>>>> Rack1R6(config-router)#do sh runn | s ip route|router ospf
>>>>>>> router ospf 1
>>>>>>> router-id 150.1.6.6
>>>>>>> log-adjacency-changes
>>>>>>> area 1 nssa
>>>>>>> redistribute static subnets
>>>>>>> network 150.1.6.6 0.0.0.0 area 1
>>>>>>> network 155.1.146.6 0.0.0.0 area 1
>>>>>>>
>>>>>>> ip route 160.1.6.0 255.255.255.0 Null0
>>>>>>>
>>>>>>>
>>>>>>> The first ABR
>>>>>>>
>>>>>>> Rack1R1#sh runn | s router ospf 1|ip route
>>>>>>>
>>>>>>> router ospf 1
>>>>>>> router-id 150.1.100.100
>>>>>>> log-adjacency-changes
>>>>>>> area 1 nssa
>>>>>>> network 150.1.1.1 0.0.0.0 area 1
>>>>>>> network 155.1.0.1 0.0.0.0 area 0
>>>>>>> network 155.1.146.1 0.0.0.0 area 1
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Rack1R1#sh ip route ospf
>>>>>>> 160.1.0.0/24 is subnetted, 1 subnets
>>>>>>> O N2 160.1.6.0 [110/20] via 155.1.146.6, 01:04:19,
>>>>>>>
>>> FastEthernet0/0
>>>
>>>>>>> 150.1.0.0/16 is variably subnetted, 4 subnets, 2 masks
>>>>>>> O 150.1.6.6/32 [110/2] via 155.1.146.6, 01:04:19,
>>>>>>>
>>> FastEthernet0/0
>>>
>>>>>>> O 150.1.5.5/32 [110/65] via 155.1.0.5, 01:05:14, Serial1/0
>>>>>>> O 150.1.4.4/32 [110/2] via 155.1.146.4, 01:04:19,
>>>>>>>
>>> FastEthernet0/0
>>>
>>>>>>>
>>>>>>>>> "Note that the route to 160.1.6.0 shows as N2"
>>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Rack1R1#sh ip ospf database | b Type-7
>>>>>>> Type-7 AS External Link States (Area 1)
>>>>>>>
>>>>>>> Link ID ADV Router Age Seq# Checksum Tag
>>>>>>> 160.1.6.0 150.1.6.6 11 0x80000004 0x005C48 0
>>>>>>>
>>>>>>> Type-5 AS External Link States
>>>>>>>
>>>>>>> Link ID ADV Router Age Seq# Checksum Tag
>>>>>>> 160.1.6.0 150.1.100.100 2037 0x80000002 0x002AC9 0
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>> This shows that Router 1 is doing the translation and is
>>>>>>>>>
>>>>>>> originating the Type-5 LSA in the OSPF domain (except Stub and NSSA
>>>>>>> areas).
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On the other ABR (R4)
>>>>>>>
>>>>>>> Rack1R4#sh ip route ospf
>>>>>>> 160.1.0.0/24 is subnetted, 1 subnets
>>>>>>> O E2 160.1.6.0 [110/20] via 155.1.146.6, 01:09:32, Ethernet0/1
>>>>>>> 150.1.0.0/16 is variably subnetted, 4 subnets, 2 masks
>>>>>>> O 150.1.6.6/32 [110/11] via 155.1.146.6, 01:09:39,
>>>>>>> Ethernet0/1
>>>>>>> O 150.1.5.5/32 [110/65] via 155.1.0.5, 01:10:23, Serial1/0
>>>>>>> O 150.1.1.1/32 [110/11] via 155.1.146.1, 01:09:39,
>>>>>>> Ethernet0/1
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>> "Note that the route to 160.1.6.0 shows as E2"
>>>>>>>>>
>>>>>>> Rack1R4#sh ip ospf data | b Type-7
>>>>>>> Type-7 AS External Link States (Area 1)
>>>>>>>
>>>>>>> Link ID ADV Router Age Seq# Checksum Tag
>>>>>>> 160.1.6.0 150.1.6.6 197 0x80000004 0x005C48 0
>>>>>>>
>>>>>>> Type-5 AS External Link States
>>>>>>>
>>>>>>> Link ID ADV Router Age Seq# Checksum Tag
>>>>>>> 160.1.6.0 150.1.100.100 180 0x80000003 0x0028CA 0
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> My questions are:
>>>>>>>
>>>>>>> 1) Why R1 is showing the route as N2 for 160.1.6.0?
>>>>>>> 2)isn't Type-5 preferred over Type-7?
>>>>>>>
>>>>>>>
>>>>>>> I would have expeceted to see R1 to have this route as E2 (i.e.
>>>>>>>
>>> similar to
>>>
>>>>>> R4).
>>>>>>
>>>>>>> Thanks,
>>>>>>> Joseph.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Nov 10, 2007 11:15 PM, Joseph Saad <joseph.samir.saad@gmail.com>
>>>>>>>
>>> wrote:
>>>
>>>>>>>> Hi Shiran,
>>>>>>>>
>>>>>>>> I am referring to IE technology Lab "NSSA Area type 7to5 LSA
>>>>>>>> Translator Election".
>>>>>>>>
>>>>>>>> Regarding your Answer:
>>>>>>>>
>>>>>>>> When you use a NSSA not so stubby area then the ASBR will generate
>>>>>>>>
>>> N2
>>>
>>>>>>>> for external into the area.
>>>>>>>>
>>>>>>>>
>>>>>>>>>> "I am OK with that"
>>>>>>>>>>
>>>>>>>> E2 is for external routes and will be generated by ASBR.
>>>>>>>>
>>>>>>>>
>>>>>>>>>> "I am OK with that for Normal Areas".
>>>>>>>>>>
>>>>>>>> in a router that have under his process an Area type NSSA will
>>>>>>>>
>>> convert
>>>
>>>>>>>> the Type 7 LSA to Type 5 when going from NSSA to the Backbone
>>>>>>>>
>>> Area.
>>>
>>>>>>>>>> "True, but what if we have multiple ABR such as this particular
>>>>>>>>>>
>>>>>>>> technology lab".
>>>>>>>>
>>>>>>>> there are no election for ABR or ASBR, if your router have under
>>>>>>>>
>>> his
>>>
>>>>>>>> process more then one area then he is ABR, if he have
>>>>>>>>
>>> redistribution
>>>
>>>>>>>> (or originating a default route) then he is ASBR.
>>>>>>>>
>>>>>>>>
>>>>>>>>>> "True, but if you have multiple ABR's for the NSSA, one of them
>>>>>>>>>>
>>>>>>>> (i.e. with the highest router-id) is elected for translating the
>>>>>>>>
>>> type
>>>
>>>>>>>> 7 into type 5 LSA".
>>>>>>>>
>>>>>>>> Joseph.
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> On Nov 10, 2007 11:00 PM, shiran guez < shiranp3@gmail.com> wrote:
>>>>>>>>
>>>>>>>>> when you use a NSSA not so stubby area then the ASBR will
>>>>>>>>>
>>> generate N2
>>>
>>>>>> for
>>>>>>
>>>>>>>>> external into the area.
>>>>>>>>>
>>>>>>>>> E2 is for external routes and will be generated by ASBR.
>>>>>>>>>
>>>>>>>>> in a router that have under his process an Area type NSSA will
>>>>>>>>>
>>> convert
>>>
>>>>>> the
>>>>>>
>>>>>>>>> Type 7 LSA to Type 5 when going from NSSA to the Backbone Area.
>>>>>>>>>
>>>>>>>>> there are no election for ABR or ASBR, if your router have under
>>>>>>>>>
>>> his
>>>
>>>>>> process
>>>>>>
>>>>>>>>> more then one area then he is ABR, if he have redistribution (or
>>>>>>>>>
>>>>>> originating
>>>>>>
>>>>>>>>> a default route) then he is ASBR.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Nov 10, 2007 8:45 PM, Joseph Saad <
>>>>>>>>>
>>> joseph.samir.saad@gmail.com>
>>>
>>>>>> wrote:
>>>>>>
>>>>>>>>>>
>>>>>>>>>> When I have 2 ABR's for an NSSA. The elected Type 7 to Type 5
>>>>>>>>>>
>>> LSA
>>>
>>>>>>>>>> translator shows the route in the routing table as N2 while
>>>>>>>>>>
>>> the
>>>
>>>>>> other
>>>>>>
>>>>>>>>>> ABR shows it as E2 route.
>>>>>>>>>>
>>>>>>>>>> I can understand why the non-elected ABR shows the route as E2
>>>>>>>>>> (because it receives both type 7 LSA and type 5 LSA and type 5
>>>>>>>>>>
>>> LSA
>>>
>>>>>>>>>> will be preferred).
>>>>>>>>>>
>>>>>>>>>> What I can't understand is why the elected router shows the
>>>>>>>>>>
>>> route as
>>>
>>>>>>>>>> N2 though it is the one which originated the type 5 LSA and
>>>>>>>>>>
>>> already
>>>
>>>>>>>>>> has the type 7 LSA in its database.
>>>>>>>>>>
>>>>>>>>>> Can someone shed some light on why such behavior?
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>> _______________________________________________________________________
>>>
>>>>>>>>>> Subscription information may be found at:
>>>>>>>>>> http://www.groupstudy.com/list/CCIELab.html
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Shiran Guez
>>>>>>>>> MCSE CCNP NCE1
>>>>>>>>> http://cciep3.blogspot.com
>>>>>>>>> http://www.linkedin.com/in/cciep3
>>>>>>>>>
>>>>>>
>>>>>> --
>>>>>>
>>>>>>
>>>>>> Shiran Guez
>>>>>> MCSE CCNP NCE1
>>>>>> http://cciep3.blogspot.com
>>>>>> http://www.linkedin.com/in/cciep3
>>>>>>
>>>>
>>>> --
>>>>
>>>>
>>>>
>>>> Shiran Guez
>>>> MCSE CCNP NCE1
>>>> http://cciep3.blogspot.com
>>>> http://www.linkedin.com/in/cciep3
>>>>
>>>
>>> --
>>>
>>>
>>> Shiran Guez
>>> MCSE CCNP NCE1
>>> http://cciep3.blogspot.com
>>> http://www.linkedin.com/in/cciep3
>>>
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Tue Jan 01 2008 - 12:04:32 ARST