OSPF External Routes (Redistributing Static Routes)

What is an OSPF external route

So far these articles have exchanged routes only inside a single OSPF domain. In a real network, routes to destinations that OSPF does not run on — a network reached by a static route, a range running another routing protocol, the internet side — also have to be carried in OSPF.

A route brought in from outside OSPF is an external route, the act of bringing it in is redistribution, and a router that does so is an ASBR (Autonomous System Boundary Router). An ASBR converts the external route into an AS External-LSA (Type 5) and floods it across the whole AS, crossing area boundaries.

Using the redistribution of a static route as the subject, this article looks at the contents of a Type 5 LSA, the difference between E1 and E2, and which route is chosen when there are two ASBRs. The details of the Forwarding Address are covered in a separate article.

Test setup

Seven Cisco IOS XR (XRd 26.1.1) routers form two independent OSPF domains. Domain A (router ospf 1) and domain B (router ospf 2) are physically joined by two links, R3–R11 and R4–R12, but no OSPF adjacency is formed on those links (they are passive enable in their own domain). In other words, domain A cannot see domain B’s routes unless they are redistributed.

There are two reasons the boundary links are in OSPF as passive enable. No adjacency is formed, so the two domains stay independent, but 172.16.3.0/24 and 172.16.4.0/24 are advertised inside each domain, so the intermediate traceroute hops have a return path and the path can be read. It also keeps the Forwarding Address at 0.0.0.0, so the article can focus on the basic behaviour of external routes.

The test runs in ten stages (STEP 0 to 9). The show output, syslog and running-config collected from all seven routers at every STEP are gathered in Verification config and show output at the end of the article.

Before redistribution (STEP 0)

R1 has not a single route from domain B. A ping to 13.13.13.13 fails, and traceroute ends with R1 itself returning !N (destination network unreachable).

RP/0/RP0/CPU0:R1#show route 13.13.13.13/32
Sun Sep  6 07:43:23.632 UTC

% Network not in table

RP/0/RP0/CPU0:R1#ping 13.13.13.13 source 1.1.1.1 count 2
Sun Sep  6 07:43:23.767 UTC
Type escape sequence to abort.
Sending 2, 100-byte ICMP Echos to 13.13.13.13 timeout is 2 seconds:
..
Success rate is 0 percent (0/2)
RP/0/RP0/CPU0:R1#traceroute 13.13.13.13 source 1.1.1.1
Sun Sep  6 07:43:28.625 UTC

Type escape sequence to abort.
Tracing the route to 13.13.13.13

 1  1.1.1.1 !N  *  !N 

At this point all R1 has learned is the routes inside domain A plus the two boundary segments.

RP/0/RP0/CPU0:R1#show route ospf
Sun Sep  6 07:40:10.467 UTC

O    2.2.2.2/32 [110/2] via 10.0.12.2, 00:00:30, GigabitEthernet0/0/0/0
O    3.3.3.3/32 [110/3] via 10.0.12.2, 00:00:30, GigabitEthernet0/0/0/0
O    4.4.4.4/32 [110/3] via 10.0.12.2, 00:00:30, GigabitEthernet0/0/0/0
O    10.0.23.0/24 [110/2] via 10.0.12.2, 00:00:30, GigabitEthernet0/0/0/0
O    10.0.24.0/24 [110/2] via 10.0.12.2, 00:00:30, GigabitEthernet0/0/0/0
O    172.16.3.0/24 [110/3] via 10.0.12.2, 00:00:30, GigabitEthernet0/0/0/0
O    172.16.4.0/24 [110/3] via 10.0.12.2, 00:00:30, GigabitEthernet0/0/0/0

Redistributing a static route to become an ASBR (STEP 1)

Configure a static route on R3 towards the destination in domain B and redistribute it into OSPF. A return path is needed too, so R11 does the same thing in the opposite direction.

router static
 address-family ipv4 unicast
  13.13.13.13/32 172.16.3.11
 !
!
router ospf 1
 redistribute static
!

router static
 address-family ipv4 unicast
  1.1.1.1/32 172.16.3.3
 !
!
router ospf 2
 redistribute static
!

Having redistributed, R3 now reports that it is an AS boundary router in show ospf.

RP/0/RP0/CPU0:R3#show ospf
Sun Sep  6 07:46:25.276 UTC

 Routing Process "ospf 1" with ID 3.3.3.3
 Role: Primary Active
 NSR (Non-stop routing) is Enabled
 Supports only single TOS(TOS0) routes
 Supports opaque LSA
 It is an autonomous system boundary router

R1’s LSDB now holds the Type 5 LSA generated by R3. The four fields we will keep coming back to in this article are all there.

RP/0/RP0/CPU0:R1#show ospf database external
Sun Sep  6 07:45:46.443 UTC


            OSPF Router with ID (1.1.1.1) (Process ID 1)

		Type-5 AS External Link States

  Routing Bit Set on this LSA
  LS age: 106
  Options: (No TOS-capability, DC)
  LS Type: AS External Link
  Link State ID: 13.13.13.13 (External Network Number)
  Advertising Router: 3.3.3.3
  LS Seq Number: 80000001
  Checksum: 0x352b
  Length: 36
  Network Mask: /32
	Metric Type: 2 (Larger than any link state path)
	TOS: 0 
	Metric: 20 
	Forward Address: 0.0.0.0
	External Route Tag: 0

It is installed in the routing table as type extern 2 and shown as O E2 by show route ospf.

RP/0/RP0/CPU0:R1#show route 13.13.13.13/32
Sun Sep  6 07:45:47.816 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 20, type extern 2
  Installed Sep  6 07:44:03.182 for 00:01:44
  Routing Descriptor Blocks
    10.0.12.2, from 3.3.3.3, via GigabitEthernet0/0/0/0
      Route metric is 20
  No advertising protos. 

Reachability works too. Traceroute takes four hops, R2 to R3 to R11 to R13, and the boundary link address 172.16.3.11 appears as a hop.

RP/0/RP0/CPU0:R1#ping 13.13.13.13 source 1.1.1.1 count 3
Sun Sep  6 07:45:10.691 UTC
Type escape sequence to abort.
Sending 3, 100-byte ICMP Echos to 13.13.13.13 timeout is 2 seconds:
!!!
Success rate is 100 percent (3/3), round-trip min/avg/max = 12/15/21 ms
RP/0/RP0/CPU0:R1#traceroute 13.13.13.13 source 1.1.1.1
Sun Sep  6 07:45:11.638 UTC

Type escape sequence to abort.
Tracing the route to 13.13.13.13

 1  10.0.12.2 6 msec  3 msec  4 msec 
 2  10.0.23.3 7 msec  22 msec  6 msec 
 3  172.16.3.11 10 msec  9 msec  9 msec 
 4  10.1.11.13 13 msec  *  18 msec 

The reverse direction is set up the same way, so R13 reaches R1 as well.

RP/0/RP0/CPU0:R13#ping 1.1.1.1 source 13.13.13.13 count 3
Sun Sep  6 07:45:18.563 UTC
Type escape sequence to abort.
Sending 3, 100-byte ICMP Echos to 1.1.1.1 timeout is 2 seconds:
!!!
Success rate is 100 percent (3/3), round-trip min/avg/max = 11/13/15 ms
RP/0/RP0/CPU0:R13#traceroute 1.1.1.1 source 13.13.13.13
Sun Sep  6 07:45:19.532 UTC

Type escape sequence to abort.
Tracing the route to 1.1.1.1

 1  10.1.11.11 6 msec  4 msec  7 msec 
 2  172.16.3.3 7 msec  9 msec  7 msec 
 3  10.0.23.2 11 msec  10 msec  10 msec 
 4  10.0.12.1 14 msec  *  21 msec 

The difference between E1 and E2 (STEP 2 and 3)

There are two ways to count the metric of an external route.

Change R3’s redistribution to metric-type 1 and compare two routers at different distances from the ASBR (R1 is cost 2 from R3, R2 is cost 1).

router ospf 1
 redistribute static metric-type 1
!

RP/0/RP0/CPU0:R1#show route 13.13.13.13/32
Sun Sep  6 07:50:04.397 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 22, type extern 1
  Installed Sep  6 07:48:50.086 for 00:01:14
  Routing Descriptor Blocks
    10.0.12.2, from 3.3.3.3, via GigabitEthernet0/0/0/0
      Route metric is 22
  No advertising protos. 

RP/0/RP0/CPU0:R2#show route 13.13.13.13/32
Sun Sep  6 07:50:27.611 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 21, type extern 1
  Installed Sep  6 07:48:50.079 for 00:01:37
  Routing Descriptor Blocks
    10.0.23.3, from 3.3.3.3, via GigabitEthernet0/0/0/1
      Route metric is 21
  No advertising protos. 

Both are looking at the same LSA (seed metric 20), yet R1 shows 22 and R2 shows 21 — 20 + 2 and 20 + 1, with the internal cost to the ASBR added.

Back on E2, both show 20.

RP/0/RP0/CPU0:R1#show route 13.13.13.13/32
Sun Sep  6 07:53:53.531 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 20, type extern 2
  Installed Sep  6 07:52:59.278 for 00:00:54
  Routing Descriptor Blocks
    10.0.12.2, from 3.3.3.3, via GigabitEthernet0/0/0/0
      Route metric is 20
  No advertising protos. 

RP/0/RP0/CPU0:R2#show route 13.13.13.13/32
Sun Sep  6 07:54:16.928 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 20, type extern 2
  Installed Sep  6 07:52:59.271 for 00:01:17
  Routing Descriptor Blocks
    10.0.23.3, from 3.3.3.3, via GigabitEthernet0/0/0/1
      Route metric is 20
  No advertising protos. 

E2 expresses “the distance beyond the AS” and is the same value wherever in the AS you look from. E1 includes the distance inside the AS, so it varies with the observation point. Which one to use is summarised in Design notes below.

How an external route is chosen

This is the heart of the article. Let us see which route is chosen when two ASBRs advertise the same external prefix.

Checking the cost to each ASBR

The “internal cost to the ASBR”, which the decision is based on, can be seen with show ospf border-routers. From R1, R3 is 2 away and R4 is 11 away (because the interface on R2 towards R4 was given a cost of 10).

RP/0/RP0/CPU0:R1#show ospf border-routers
Sun Sep  6 07:58:29.441 UTC

OSPF 1 Internal Routing Table

Codes: i - Intra-area route, I - Inter-area route

i 3.3.3.3 [2] via 10.0.12.2, GigabitEthernet0/0/0/0, ASBR , Area 0, SPF 8
i 4.4.4.4 [11] via 10.0.12.2, GigabitEthernet0/0/0/0, ASBR , Area 0, SPF 8

The STEPs that follow use this gap — 2 to R3, 11 to R4 — to separate the decision rules.

When there are two E2 routes (STEP 4)

Have R4 redistribute the same prefix as R3, with the same seed metric 20 and the same E2.

router static
 address-family ipv4 unicast
  13.13.13.13/32 172.16.4.12
 !
!
router ospf 1
 redistribute static metric 20
!

Two Type 5s now sit in the LSDB. Only the Advertising Router differs; the metrics are the same.

RP/0/RP0/CPU0:R1#show ospf database external
Sun Sep  6 07:58:28.845 UTC


            OSPF Router with ID (1.1.1.1) (Process ID 1)

		Type-5 AS External Link States

  Routing Bit Set on this LSA
  LS age: 332
  Options: (No TOS-capability, DC)
  LS Type: AS External Link
  Link State ID: 13.13.13.13 (External Network Number)
  Advertising Router: 3.3.3.3
  LS Seq Number: 80000003
  Checksum: 0x312d
  Length: 36
  Network Mask: /32
	Metric Type: 2 (Larger than any link state path)
	TOS: 0 
	Metric: 20 
	Forward Address: 0.0.0.0
	External Route Tag: 0

  LS age: 96
  Options: (No TOS-capability, DC)
  LS Type: AS External Link
  Link State ID: 13.13.13.13 (External Network Number)
  Advertising Router: 4.4.4.4
  LS Seq Number: 80000001
  Checksum: 0x1745
  Length: 36
  Network Mask: /32
	Metric Type: 2 (Larger than any link state path)
	TOS: 0 
	Metric: 20 
	Forward Address: 0.0.0.0
	External Route Tag: 0

The seed metrics tie, so R3, the one with the smaller cost to the ASBR (2 < 11), is chosen. Traceroute goes via R3 too.

RP/0/RP0/CPU0:R1#show route 13.13.13.13/32
Sun Sep  6 07:58:01.829 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 20, type extern 2
  Installed Sep  6 07:52:59.278 for 00:05:02
  Routing Descriptor Blocks
    10.0.12.2, from 3.3.3.3, via GigabitEthernet0/0/0/0
      Route metric is 20
  No advertising protos. 
RP/0/RP0/CPU0:R1#traceroute 13.13.13.13 source 1.1.1.1
Sun Sep  6 07:58:01.959 UTC

Type escape sequence to abort.
Tracing the route to 13.13.13.13

 1  10.0.12.2 33 msec  4 msec  5 msec 
 2  10.0.23.3 8 msec  7 msec  7 msec 
 3  172.16.3.11 42 msec  10 msec  9 msec 
 4  10.1.11.13 13 msec  *  16 msec 

When the E2 seed metrics differ (STEP 5)

Lower R4’s seed metric to 15. R4 is at a disadvantage with a cost of 11 to the ASBR, but E2 does not use that internal cost in the comparison.

router ospf 1
 redistribute static metric 15
!

RP/0/RP0/CPU0:R1#show route 13.13.13.13/32
Sun Sep  6 08:02:12.442 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 15, type extern 2
  Installed Sep  6 08:01:27.256 for 00:00:45
  Routing Descriptor Blocks
    10.0.12.2, from 4.4.4.4, via GigabitEthernet0/0/0/0
      Route metric is 15
  No advertising protos. 
RP/0/RP0/CPU0:R1#traceroute 13.13.13.13 source 1.1.1.1
Sun Sep  6 08:02:12.635 UTC

Type escape sequence to abort.
Tracing the route to 13.13.13.13

 1  10.0.12.2 7 msec  6 msec  44 msec 
 2  10.0.24.4 8 msec  7 msec  7 msec 
 3  172.16.4.12 10 msec  8 msec  9 msec 
 4  10.1.12.13 14 msec  *  14 msec 

The path switched to R4, and the traceroute hops from the second one on changed to 10.0.24.4 to 172.16.4.12 to 10.1.12.13.

This is the important part. If the internal cost were added for the comparison, R3 would be 20 + 2 = 22 and R4 would be 15 + 11 = 26, so R3 should win. R4 was chosen instead, which confirms that only the seed metric is used when comparing E2 routes.

When E1 and E2 are mixed (STEP 6)

Put only R3 back to metric-type 1 (E1, seed metric 20). R3’s metric becomes 20 + 2 = 22, a larger value than R4’s E2 metric of 15.

router ospf 1
 redistribute static metric 20 metric-type 1
!

RP/0/RP0/CPU0:R1#show route 13.13.13.13/32
Sun Sep  6 08:06:15.782 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 22, type extern 1
  Installed Sep  6 08:05:29.579 for 00:00:46
  Routing Descriptor Blocks
    10.0.12.2, from 3.3.3.3, via GigabitEthernet0/0/0/0
      Route metric is 22
  No advertising protos. 
RP/0/RP0/CPU0:R1#traceroute 13.13.13.13 source 1.1.1.1
Sun Sep  6 08:06:15.990 UTC

Type escape sequence to abort.
Tracing the route to 13.13.13.13

 1  10.0.12.2 6 msec  5 msec  5 msec 
 2  10.0.23.3 8 msec  7 msec  7 msec 
 3  172.16.3.11 10 msec  9 msec  9 msec 
 4  10.1.11.13 12 msec  *  17 msec 

The metric values are 22 (R3) against 15 (R4), so R4’s is smaller, yet R3 was chosen. When comparing external routes, the metric type decides the preference before the metric is even looked at. E1 is always preferred over E2.

When there are two E1 routes (STEP 7)

Set R4 to metric-type 1 as well (seed metric 15). Both are E1 now, so this time the metrics are compared.

• R3: 20 + 2 = 22
• R4: 15 + 11 = 26

router ospf 1
 redistribute static metric 15 metric-type 1
!

RP/0/RP0/CPU0:R1#show route 13.13.13.13/32
Sun Sep  6 08:10:17.038 UTC

Routing entry for 13.13.13.13/32
  Known via "ospf 1", distance 110, metric 22, type extern 1
  Installed Sep  6 08:05:29.579 for 00:04:47
  Routing Descriptor Blocks
    10.0.12.2, from 3.3.3.3, via GigabitEthernet0/0/0/0
      Route metric is 22
  No advertising protos. 

Traceroute stays on the path via R3.

RP/0/RP0/CPU0:R1#traceroute 13.13.13.13 source 1.1.1.1
Sun Sep  6 08:10:17.805 UTC

Type escape sequence to abort.
Tracing the route to 13.13.13.13

 1  10.0.12.2 7 msec  5 msec  5 msec 
 2  10.0.23.3 8 msec  8 msec  7 msec 
 3  172.16.3.11 11 msec  10 msec  9 msec 
 4  10.1.11.13 13 msec  *  17 msec 

R4 has the smaller seed metric (15 < 20) but loses on the total, so R3 was chosen.

Putting STEP 5 and STEP 7 side by side makes it clear. The seed metrics are exactly the same pair, 20 and 15, yet E2 picks R4 and E1 picks R3 — the opposite result.

Identical conditions produce ECMP (the domain B side of STEP 4)

The domain B side is built the same way. Seen from R13, R11 and R12 advertise 1.1.1.1/32 with the same E2 and the same seed metric 20, and the internal cost from R13 to both ASBRs is the same, 1. Neither can be preferred, so both go into the routing table (equal-cost multipath).

RP/0/RP0/CPU0:R13#show route 1.1.1.1/32
Sun Sep  6 08:00:52.365 UTC

Routing entry for 1.1.1.1/32
  Known via "ospf 2", distance 110, metric 20, type extern 2
  Installed Sep  6 07:57:10.361 for 00:03:42
  Routing Descriptor Blocks
    10.1.11.11, from 11.11.11.11, via GigabitEthernet0/0/0/0
      Route metric is 20
    10.1.12.12, from 12.12.12.12, via GigabitEthernet0/0/0/1
      Route metric is 20
  No advertising protos. 

Summary of the selection rules

The order confirmed here is not something Cisco’s implementation decided on its own: it is specified in RFC 2328 Section 16.4, step (6). The RFC states that (a) through (d) are evaluated in order.

Paths judged to be of equal preference are simply added to the entry’s list of paths, as stated at the top of step (6). That is the equal-cost multipath observed on the domain B side.

The following values, on the other hand, are not specified by the RFC and are up to the implementation. They may differ on another vendor or another release, so confirm them on the real device when designing.

Originating a default route (STEP 8)

Instead of redistributing individual prefixes, a default route (0.0.0.0/0) can be advertised as a Type 5. The ASBR effectively says “send anything outside the AS to me”.

router ospf 1
 default-information originate always
!

always is there to advertise it even when R3 itself has no default route. Without it, the default is advertised only while R3’s own routing table has one.

RP/0/RP0/CPU0:R1#show route ospf
Sun Sep  6 08:14:25.707 UTC

O*E2 0.0.0.0/0 [110/1] via 10.0.12.2, 00:00:46, GigabitEthernet0/0/0/0
O    2.2.2.2/32 [110/2] via 10.0.12.2, 00:34:45, GigabitEthernet0/0/0/0
O    3.3.3.3/32 [110/3] via 10.0.12.2, 00:34:45, GigabitEthernet0/0/0/0
O    4.4.4.4/32 [110/12] via 10.0.12.2, 00:17:45, GigabitEthernet0/0/0/0
O    10.0.23.0/24 [110/2] via 10.0.12.2, 00:34:45, GigabitEthernet0/0/0/0
O    10.0.24.0/24 [110/11] via 10.0.12.2, 00:17:45, GigabitEthernet0/0/0/0
O E1 13.13.13.13/32 [110/22] via 10.0.12.2, 00:08:56, GigabitEthernet0/0/0/0
O    172.16.3.0/24 [110/3] via 10.0.12.2, 00:34:45, GigabitEthernet0/0/0/0
O    172.16.4.0/24 [110/12] via 10.0.12.2, 00:17:45, GigabitEthernet0/0/0/0
RP/0/RP0/CPU0:R1#show route 0.0.0.0/0
Sun Sep  6 08:14:25.849 UTC

Routing entry for 0.0.0.0/0
  Known via "ospf 1", distance 110, metric 1, candidate default path
  Tag 1, type extern 2
  Installed Sep  6 08:13:39.774 for 00:00:46
  Routing Descriptor Blocks
    10.0.12.2, from 3.3.3.3, via GigabitEthernet0/0/0/0
      Route metric is 1
  No advertising protos. 

The * in O*E2 marks it as a default route candidate. The metric is 1, a different default from the 20 used by redistribute, which is worth noting. show route 0.0.0.0/0 also reports candidate default path.

This output still carries the results of the earlier STEPs: O E1 13.13.13.13/32 [110/22] is the E1 external route chosen in STEP 7, and 4.4.4.4/32 at [110/12] and 10.0.24.0/24 at [110/11] reflect the cost 10 added in STEP 4.

Design notes

• E2 (the default) expresses only “the distance to outside the AS”. Because the value is the same from anywhere in the AS, it suits the case where you want the internal distance to decide which of several ASBRs traffic leaves through (the behaviour in STEP 4).
• Use E1 when you want to compare on the total cost including the distance inside the AS. Note, however, that when E1 and E2 are mixed E1 wins regardless of the metric values (STEP 6), so keep the metric type consistent if several ASBRs redistribute the same prefix. Setting only one of them to E1 quietly pulls everything to that one.
• The default seed metric is 20. If several ASBRs redistribute with the default, the choice comes down to the internal cost to the ASBR, as in STEP 4. Set it explicitly when you want a particular split.
• Watch for loops when redistributing in both directions. It is not a problem here because each domain redistributes only static routes, but mutually redistributing OSPF on both sides can send routes back and forth and destabilise them.
• An ASBR keeps using the route it redistributed from. In R3’s routing table the static route (administrative distance 1) beats the OSPF external route (110), so the external route is not installed. Observation has to be done on a router other than the ASBR.

About the Forwarding Address

In every STEP here, the Forward Address of the Type 5 LSA was 0.0.0.0, which means “send traffic for this external route to the ASBR itself”.

Under the right conditions an ASBR puts an address that is not its own there, and traffic then goes straight to that next hop without passing through the ASBR. The conditions and the behaviour are covered in a separate article.

Verification config and show output

At every STEP the following three files were collected from all seven routers, split per router. The verification config is the ..._run.txt file (the final state is the one from STEP 9).

Reachability checks (ping / traceroute) were taken as separate ..._trace.txt files on R1 and R13 only, where they mean something.

STEP 0: initial state (no redistribution) — R1 does not know domain B’s routes; ping fails

Additional: R1 routes and ping / R13 routes and ping

STEP 1: redistribute statics on R3 and R11 in both directions — a Type 5 appears; ping / traceroute succeed both ways

Additional: R1 routes and ping/traceroute / R13 routes and ping/traceroute

STEP 2: set metric-type 1 (E1) on R3 — R1 shows 22 and R2 shows 21 (the internal cost is added)

Additional: R1 routes and traceroute

STEP 3: put R3 back to E2 — both R1 and R2 show 20

STEP 4: cost 10 on R2 towards R4, redistribute on R4 and R12 (E2, 20) — two Type 5s; R3 with a cost of 2 is chosen

Additional: R1 cost to the ASBRs and traceroute / R13 routes and traceroute

STEP 5: lower R4’s seed metric to 15 — the path switches to R4 (only the seed metric is compared)

Additional: R1 routes and traceroute

STEP 6: set metric-type 1 (E1, 20) on R3 — E1 wins and the path returns to R3

Additional: R1 routes and traceroute

STEP 7: set metric-type 1 (E1, 15) on R4 as well — 22 < 26 on the totals, still via R3

Additional: R1 routes, the Type 5s and traceroute

STEP 8: default-information originate always on R3 — O*E2 0.0.0.0/0 is advertised

Additional: R1 the default route

STEP 9: remove the redistribution, statics and cost setting (final state) — restored to the initial state

Additional: R1 confirming the rollback

References

Related articles

• What Is OSPF
• OSPF Router ID
• OSPF Packet Types and Header Format
• OSPF Neighbor States
• OSPF Options Field
• DR and BDR
• OSPF Network Types