loopback not being advertised through inter area OSPF

[francisputi]

if link from CHR gi0/1 is UP loopback 1 from CER02 is being advertised through area0 but if that link is down CER01 is not able to see the route 2.2.2.2 but the PB can see it through InterArea. may i know the reason why?

CER01

interface Loopback1
ip address 1.1.1.1 255.255.255.255
ip ospf 1 area 0
!
interface GigabitEthernet0/0
ip address 10.10.0.1 255.255.255.248
ip ospf 1 area 16
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/1
no ip address
shutdown
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/2
no ip address
shutdown
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/3
ip address 10.100.0.1 255.255.255.248
ip ospf 1 area 0
duplex auto
speed auto
media-type rj45
!
router ospf 1

CER02

interface Loopback1
ip address 2.2.2.2 255.255.255.255
ip ospf 1 area 0
!
interface GigabitEthernet0/0
ip address 10.200.0.1 255.255.255.248
ip ospf 1 area 0
duplex auto
speed auto
!
interface GigabitEthernet0/1
ip address 10.10.1.1 255.255.255.248
ip ospf 1 area 16
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/2
no ip address
shutdown
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/3
no ip address
shutdown
duplex auto
speed auto
media-type rj45
!
router ospf 1
!

PB

interface GigabitEthernet0/0
ip address 10.10.0.2 255.255.255.248
ip ospf 1 area 16
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/1
ip address 10.10.1.2 255.255.255.248
ip ospf 1 area 16
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/2
no ip address
shutdown
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/3
no ip address
shutdown
duplex auto
speed auto
media-type rj45
!
router ospf 1

CHR

interface GigabitEthernet0/0
ip address 10.100.0.2 255.255.255.248
ip ospf 1 area 0
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/1
ip address 10.200.0.2 255.255.255.248
ip ospf 1 area 0
shutdown
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/2
no ip address
shutdown
duplex auto
speed auto
media-type rj45
!
interface GigabitEthernet0/3
no ip address
shutdown
duplex auto
speed auto
media-type rj45
!
router ospf 1
!

[Georg Pauwen]

Hello,

at first glance, looks like when that link is down, you have a partitioned backbone area 0, which means, there is an interface (the loopback2 with IP adress 2.2.2.2) on CER02 that does not have a direct link to area 0 on CHR. Area 16 is 'in the way', so to speak.

[Giuseppe Larosa]

Hello @francisputi ,

>> if link from CHR gi0/1 is UP loopback 1 from CER02 is being advertised through area0 but if that link is down CER01 is not able to see the route 2.2.2.2 but the PB can see it through InterArea. may i know the reason why?

as @Georg Pauwen has noted and we can see this in your network diagram when the CHR link gi0/1 is shiut down you are actually breaking the original area 0 in two parts.

Being advertised in area 0 in the loopback of CER02, the router would like to send a Router LSA in area 0 but there is no physical link to send it out.

The best solution for a partitioned backbone area is to add redundancy if you add a link between CER01 ad CER02 in area 0 the single link failure is not an issue anymore.

A short term solution would be the use of a virtual link going via area 16 between CER02 and CER01. The virtual link would create a logical link in area 0 via the transit area 16 allowing for the CR02 router LSA in area 0 to be sent over it.

But the forwarding plane happens in area 16 that needs to be a standard area to avoid to have traffic black holiing.

It is not somethiing to use in real networks just a tool to play with in labs.

In real world each router has / should have  at least two different links to connect to area 0

Edit:

on PB check the OSPF database to see who created the LSA type 3 for 2.2.2.2/32 it may be a relic from when CER02 was connected to area 0.

show ip ospf database summary 2.2.2.2

look for the advertising Router ID field.

if it is generated by 2.2.2.2 itself it is a relic and so on the long term also PB will miss a route for 2.2.2.2/32.

Hope to help

Giuseppe

[francisputi]

hello @Giuseppe Larosa 

here is the result when link from CHR and CER 01 is up

PB#sho ip ospf database summary 2.2.2.2

OSPF Router with ID (10.10.0.2) (Process ID 1)

Summary Net Link States (Area 16)

LS age: 1435
Options: (No TOS-capability, DC, Upward)
LS Type: Summary Links(Network)
Link State ID: 2.2.2.2 (summary Network Number)
Advertising Router: 1.1.1.1
LS Seq Number: 80000002
Checksum: 0x2B02
Length: 28
Network Mask: /32
MTID: 0 Metric: 3

LS age: 1673
Options: (No TOS-capability, DC, Upward)
LS Type: Summary Links(Network)
Link State ID: 2.2.2.2 (summary Network Number)
Advertising Router: 2.2.2.2
LS Seq Number: 80000006
Checksum: 0xF036
Length: 28
Network Mask: /32
MTID: 0 Metric: 1

here is the result when that link is down (my question here is that why PB is able to received the 2.2.2.2 as an interArea routes while CER01 is not able to? as the PB received that route from CER as an IA should it advertised it to its own area? or is a loop prevention feature why its not advertising it to CER01

PB#sho ip ospf database summary 2.2.2.2

OSPF Router with ID (10.10.0.2) (Process ID 1)

Summary Net Link States (Area 16)

LS age: 1708
Options: (No TOS-capability, DC, Upward)
LS Type: Summary Links(Network)
Link State ID: 2.2.2.2 (summary Network Number)
Advertising Router: 2.2.2.2
LS Seq Number: 80000006
Checksum: 0xF036
Length: 28
Network Mask: /32
MTID: 0 Metric: 1

PB#sho ip route
PB#sho ip route
Codes: L - local, C - connected, S - static, R - RIP, M - mobile, B - BGP
D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
ia - IS-IS inter area, * - candidate default, U - per-user static route
o - ODR, P - periodic downloaded static route, H - NHRP, l - LISP
a - application route
+ - replicated route, % - next hop override, p - overrides from PfR

Gateway of last resort is not set

1.0.0.0/32 is subnetted, 1 subnets
O IA 1.1.1.1 [110/2] via 10.10.0.1, 06:08:50, GigabitEthernet0/0
2.0.0.0/32 is subnetted, 1 subnets
O IA 2.2.2.2 [110/2] via 10.10.1.1, 03:15:51, GigabitEthernet0/1
10.0.0.0/8 is variably subnetted, 6 subnets, 2 masks
C 10.10.0.0/29 is directly connected, GigabitEthernet0/0
L 10.10.0.2/32 is directly connected, GigabitEthernet0/0
C 10.10.1.0/29 is directly connected, GigabitEthernet0/1
L 10.10.1.2/32 is directly connected, GigabitEthernet0/1
O IA 10.100.0.0/29 [110/2] via 10.10.0.1, 06:08:50, GigabitEthernet0/0
O IA 10.200.0.0/29 [110/2] via 10.10.1.1, 03:14:05, GigabitEthernet0/1
PB#

thank you

[Giuseppe Larosa]

Hello @francisputi ,

Actually PB sees an LSA that is generated by CER02 2.2.2.2.

>>>PB#sho ip ospf database summary 2.2.2.2

OSPF Router with ID (10.10.0.2) (Process ID 1)

Summary Net Link States (Area 16)

>>>>> LS age: 1708
Options: (No TOS-capability, DC, Upward)
LS Type: Summary Links(Network)
Link State ID: 2.2.2.2 (summary Network Number)
Advertising Router: 2.2.2.2
>>>> LS Seq Number: 80000006
Checksum: 0xF036
Length: 28
Network Mask: /32
MTID: 0 Metric: 1

It would be interesting to verify if the LSA is refreshed by CER02 ,that its LS age should never become grater then 1800 seconds and the sequence number should increase by 1 every 30 minutes at each refresh. Currently it is 80000006 next value shoild be 80000007.

If this happens means that CER02 considers itself still an ABR router and that just having the loopback in area 0 is enough to  make it think it is connected to area 0 ( an ABR works if it connected to area 0).

if the LSA is not refreshed it will age out when its LS age reaches 3600 seconds one hour and also the route will be removed.

I would like to see a feedback about this.  It is an interesting aspect of the choice of putting the loopback in area 0.

To be noted OSPF knows what a loopback is and in fact if you try to debug ospf hello packets on the loopback OSPF will not send out them . On the other hand EIGRP and IS-IS actually send hellos on the loopback receive them back and complain about this. This is the reason why we see the loopback configured as passive interface in those protocols.

Hope to help

Giuseppe

[francisputi]

hello @Giuseppe Larosa 

here is the result, routes was not removed as the LSA is still being refreshed

PB#sho ip ospf database summary 2.2.2.2

OSPF Router with ID (10.10.0.2) (Process ID 1)

Summary Net Link States (Area 16)

LS age: 1622
Options: (No TOS-capability, DC, Upward)
LS Type: Summary Links(Network)
Link State ID: 2.2.2.2 (summary Network Number)
Advertising Router: 2.2.2.2
LS Seq Number: 8000000C
Checksum: 0xE43C
Length: 28
Network Mask: /32
MTID: 0 Metric: 1

 It is an interesting aspect of the choice of putting the loopback in area 0??? is this not recommended by cisco?

PB#sho ip ospf database summary 2.2.2.2

OSPF Router with ID (10.10.0.2) (Process ID 1)

Summary Net Link States (Area 16)

LS age: 303
Options: (No TOS-capability, DC, Upward)
LS Type: Summary Links(Network)
Link State ID: 2.2.2.2 (summary Network Number)
Advertising Router: 2.2.2.2
LS Seq Number: 8000000D
Checksum: 0xE23D
Length: 28
Network Mask: /32
MTID: 0 Metric: 1

PB#

[Giuseppe Larosa]

Hello @francisputi ,

thanks for your feedback.

the LSA is refreshed so CER02 considers itself a valid ABR connected to area 0 even if the only interface in area 0 is its loopback.

>> It is an interesting aspect of the choice of putting the loopback in area 0??? is this not recommended by cisco?

On an ABR I would put the loopback under area 0, but your lab shows us that an ABR needs to be connected to area 0 via at least two physical links ( or logical subiinterfces but referring to different main interfaces ) so that a single link failure does not isolate its loopback.

In this moment CER02 is known only within area 16 on internal routers of area 16. CER01 that has a valid connection to area 0 receives the same summary LSAs for prefix 2.2.2.2/32 but it ignores them as they are not coming from area 0.

So 2.2.2.2 is unreachable from area 0 and all other areas. As CER01 does not re-generate in area 0 the ignored LSA type 3.

Adding a link between CER01 and CER02 and putting it in area 0 solves this issue.

Hope to help

Giuseppe

[MHM Cisco World]

Are both loopback config with area 0 or other area ?

[francisputi]

hello @MHM Cisco World 

yes on both CER looback is configured on the area0

[MHM Cisco World]

https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/iproute_ospf/configuration/xe-3e/iro-xe-3e-book/iro-mode-ospfv2.pdf

[Elliot Dierksen]

Adding this command to the loopback interface in question will make OSPF advertise it.

ip ospf network point-to-point

[paul driver]

Hello
Just like to add when advertising any loopback interfaces you need to make sure they advertised with the correct subnet mask, By default loopbacks don’t form ospf peering’s so they are classed as stub networks and will be seen as single hosts, So any loopback interface which is assigned a network mask other than an host address (/32) ospf wont advertised correctly, To rectify this change the loopback interface ospf network type to be point-to point and the correct network mask should then be advertised correctly.(as stated by @Elliot Dierksen)

Int loX
ip ospf network point-to-point