Routing

Hi All

 

I have a problem on OSPF backup path.

 

My router HKG02 only is shown one link to JP01. But HKG01 will be shown two links to JP01.

If destination is KOREA02, HKG02 have two links , and HKG01 have onyl one.

I am sure OSPF neighbor is fine as attachment. Can you guys give some direction for me to invesgate ? 

 

Or just OSPF loop prevention make it ? Is there any command to display this information ?

Thanks 

Cloud

 

#ospf

 You are using LFA right?

The HKG02 has a direct path to reach JP01 and it seems like, HKG01 uses HKG02 to reach JP01 hence LFA is failing on HKG02.

Top prove this, you can paste “show route 172.16.128.2” on HKG01 and check if it’s using HKG02 as a next-hop. Also “show ospf backup neighbors” and “show ospf backup lsp” from both HKG01 and HKG02?

Hi Kingman

 

Yes, it's LFA.

 

** HGK01 used HKG02 as next-hop ,right.

 

HKG01_RE1> show route 172.16.128.2

inet.0: 691646 destinations, 3524041 routes (691387 active, 11 holddown, 1033 hidden)

+ = Active Route, - = Last Active, * = Both

 

172.16.128.2/32    *[LDP/9] 03:56:25, metric 6

                    > to 192.168.146.37 via ae8.1, Push 302432

                      to 192.168.147.33 via xe-1/3/2.1, Push 318768

                    [OSPF/10] 03:58:05, metric 6

                    > to 192.168.146.37 via ae8.1

                      to 192.168.147.33 via xe-1/3/2.1

 

** But using the command you suggested, seems not be the same as what you said.

Did I set anything worng ? Or it is normal ? 

 

HKG01_RE1> show ospf backup neighbor
Topology default backup neighbors:

Area 0.0.0.0 backup neighbors:

192.168.146.6 via: Direct
Neighbor to Self Metric: 1
Self to Neighbor Metric: 1
Direct next-hop: ae8.1 via 192.168.146.37

 

HKG01_RE1> show ospf backup lsp

{master}

 

HKG01_RE1> show ospf route 172.16.128.2
Topology default Route Table:

Prefix Path Route NH Metric NextHop Nexthop
Type Type Type Interface Address/LSP
172.16.128.2 Intra AS BR IP 6 ae8.1 192.168.146.37
Bkup IP xe-1/3/2.1 192.168.147.33
172.16.128.2/32 Intra Network IP 6 ae8.1 192.168.146.37
Bkup IP xe-1/3/2.1 192.168.147.33

 

HKG02_RE1> show ospf backup neighbor
Topology default backup neighbors:

192.168.146.2 via: Direct
Neighbor to Self Metric: 1
Self to Neighbor Metric: 1
Direct next-hop: ae8.1 via 192.168.146.38

 

172.16.128.2 via: Direct
Neighbor to Self Metric: 5
Self to Neighbor Metric: 5
Direct next-hop: xe-1/2/2.0 via 172.16.128.41

 

HKG02_RE1> show ospf backup lsp

{master}

 

HKG02_RE1> show ospf route 172.16.128.2
Topology default Route Table:

Prefix Path Route NH Metric NextHop Nexthop
Type Type Type Interface Address/LSP
172.16.128.2 Intra AS BR IP 5 xe-1/2/2.0 172.16.128.41
172.16.128.2/32 Intra Network IP 5 xe-1/2/2.0 172.16.128.41

Can you please explain this?

 

** But using the command you suggested, seems not be the same as what you said.

 

The outputs looks normal to me. If HKG01 is using as a next-hop to HKG02 to reach JP01, HKG02 can’t use HKG01 as a LFA to reach JP01.

Hi Kingsman

 

Sorry to explain detail not enough.

 

I mean "show ospf backup neighbor" is fine, but should not display the backup neighbor on both HK router.

Or it is normal as what you said it looks like be fine.

 

Moreover, how to prove it is fine ? should have some command to display the hidden backup path or neighbor ?

 

Thanks for your patience. 

 

Best regards,

Cloud 

Hi All

 

I think problem is on why HK routers don't exchange some LFA.

 

For example, HKG02 have backup path to KOREA01 from JP01--> KOREA02.

But HKG02 don't share this path to HKG01.Thus HKG01 only have one path what connect directly to KOREA01.

 

Is there any way to figureo out this problem ?

 

HKG02> show ospf backup spf detail 192.168.144.2
Topology default results:

Area 0.0.0.0 results:

192.168.144.2
Self to Destination Metric: 6
Parent Node: 192.168.146.2
Primary next-hop: ae8.1 via 192.168.146.38
Backup Neighbor: 103.198.146.2 via: Direct
Neighbor to Destination Metric: 5, Neighbor to Self Metric: 1
Self to Neighbor Metric: 1, Backup preference: 0x0
Not eligible, Reason: Primary next-hop link fate sharing
Path:192.168.146.2;192.168.144.2;

 

Backup Neighbor: 172.16.128.2 via: Direct
Neighbor to Destination Metric: 6, Neighbor to Self Metric: 5
Self to Neighbor Metric: 5, Backup preference: 0x0
Eligible, Reason: Contributes backup next-hop
Path:172.16.128.2;192.168.144.6;192.168.144.2;

Hi,

I already explaind it.

You have to look at the topology, if HKG01’s both neighbor (HKG02 and TPE01) uses HKG01 as a next-hop to reach the KOREA01, the LFA will fail.

LFA is not suitable for the ring type of topology, that’s why rLFA was developed.

You can try rLFA and check

Hi Kingsman

 

Yes, HKG01's both neighbors (HKG02/TPE01) next-hop is HKG01.

Can you explani why LFA will be failed ?

 

Also,I found wired log from OSPF in HKG02 from neghbors like below information :

 

May 8 15:25:35.812138 Walk SPF: Node attributes copied 192.168.145.2
May 8 15:25:35.812140 Walk SPF: Add candidate node node 172.16.128.2
May 8 15:25:35.812141 Inheriting parent attributes Parent 192.168.146.6
May 8 15:25:35.812144 Match link parameter:
May 8 15:25:35.812146 Link-id doesn't match topo-link 172.16.128.2,TED entry 0.0.0.0
May 8 15:25:35.812149 Opaque LSA not found :
May 8 15:25:35.812153 link not found type PointToPoint link-id 172.16.128.2 link data 10.7.100.42 adding color zero

 

Does it relate to the ring topolgy what you said it's not suitabe ?

 

And rLFA is remote LFA ?

 

Many thanks 

Cloud

Hi,

 

yes, rLFA is remote LFA.

 

To underatand LFA, you can refer this excellent blog. It will hellp you understand the LFA technology and why  LFA is failing in your topology.

 

http://packetpushers.net/ip-frr-micro-loops-part-1/

 

For the logs, i would suggest you to please open a new thread.