OSPF Metric

[hs08]

Hello,

I have topology below.

From R1 perspective when i issue show ip route 10.103.248.0 i got

Routing entry for 10.103.248.0/24
Known via "ospf 100", distance 110, metric 43, type inter area
Last update from 10.102.200.2 on GigabitEthernet1/0/2, 1d06h ago
Routing Descriptor Blocks:
10.102.200.2, from 10.103.103.5, 1d06h ago, via GigabitEthernet1/0/2
Route metric is 43, traffic share count is 1
* 10.102.173.10, from 10.103.103.5, 1d22h ago, via GigabitEthernet1/0/1
Route metric is 43, traffic share count is 1

The route metric for both path is 43, this metric is cost from R1 to R2 and from R1 to R3? If true, why the metric is 43 and not 2? The link between R1 to R2 and R3 is 1Gbps.

[paul driver]

Hello

---

@hs08 wrote:

Routing entry for 10.103.248.0/24
Known via "ospf 100", distance 110, metric 43, type inter area

---

The cost is the forward metric to reach the ABR of the area that is adverting the prefix

[hs08]

Can you explain more specific? I already add the ospf area in the picture.

[MHM Cisco World]

The route metric for both path is 43, this metric is cost from R1 to R2 and from R1 to R3?

No

From R1 to R2/R3 plus R2/R3 to R4/R5 plus R4/R5 to R6 plus link connect to R6

It cost for all link along path. 

MHM

[hs08]

Ok noted, 

i can see the metric from R5 to 10.103.248.0 is 41 this mean metric from R1 become 43 due 42 plus R3 to R5 plus R3 to R1, am i correct?

Now the question is why metric from R5 to 10.103.248.0 is 41?

this from R5

Routing entry for 10.103.248.0/24
Known via "ospf 100", distance 110, metric 41, type intra area
Last update from 10.103.245.9 on GigabitEthernet0/0/0, 7w0d ago
Routing Descriptor Blocks:
* 10.103.245.9, from 10.103.103.1, 7w0d ago, via GigabitEthernet0/0/0
Route metric is 41, traffic share count is 1

here R5 config for int g0/0/0

interface GigabitEthernet0/0/0
ip address 10.103.245.10 255.255.255.252
ip ospf network point-to-point
ip ospf 100 area 0.0.0.0
negotiation auto

And here config on R6

interface Ethernet2/17
ip address 10.103.245.9/30
ip ospf network point-to-point
ip router ospf 100 area 0.0.0.0
no shutdown

[MHM Cisco World]

Friend

I already share command with small lab in your other post

You need to use 

Show ip ospf database route <- for intra

To see cost for each link

Show run don't give you anything 

MHM

[hs08]

Noted

[Giuseppe Larosa · Accepted solution]

Hello @hs08 ,

is R6 a Nexus running NX OS ?

they have a different default reference bandwidth for OSPF it is not the same value used by IOS XE or IOS .

you can check this with

show ip ospf

Hope to help

Giuseppe

[hs08]

Hi..

Yes correct, when i check on the documentation the NX-OS use 40Gbps for bandwidth reference, so R5 will have metric 41. This make me understand from R1 the metrcis is 43.

[MHM Cisco World]

MHM

[hs08]

Hi..

From R5 perspective the cost to 10.103.0.0/16 is 41, so there will be 2 costs, one from R5 to R6 and R6 to the host?

NX-OS use 40Gbps for reference bandwidth, this mean :
Cost from R5 to R6 is 4, and from R6 to the host is 1? If true, why cost from R6 to host is not 4? In my mind cost from R5 to R6 will same with R6 to the host, so the total cost should be (40 + 40 = 80).

[paul driver]

Hello

---

@hs08 wrote:

Can you explain more specific? I already add the ospf area in the picture.

---

R6 suggest its an ABR between Area0 and another area, as such a type ospf 3 lsa will be advertised  into Area0 by R6 for that prefix its adverting
OSPF rtrs in area 0 will then have forward metric calculation based on the route path to reach R6 ABR 

[paul driver]

Hello
LOL - apologies my mistake your ABRs are both  R4/R5, so the calculation is the advertised type 3 LSA  from either R4/R5 (41)and the ospf metric to reach both R4/R5 (2)