Route Conflict - OSPF


#1

Hi!

I have a strange behaviour:

Balance One is connected to two hubs - Primary (costs 10) and DR (costs 20).
If I reboot the Primary hub, PepVPN-Status shows: Route Conflict.

The goal is to have a hub-redundancy.

Can you give me a hint, what happens? The tunnel that shows “route conflicts” does handle traffic!

I think, I have a problem with OSPF.
The only thing, I have changed is: I added the LAN-interface to Area 0.0.0.0, as on the LAN-Interface, there is a second router that need to know the OSPF-routes of the two Peplink-Hubs (Balance 710).

Thank you for your help!


#2

Hello KPS,

Usually ‘Route Conflict’ means that two routers have the same IP address range or overlapping IP ranges. Please check your DHCP Settings to make sure that the routers aren’t overlapping.


#3

Hi Zach!

DHCP is not activated, but my two “head-quarter balance 710” have both one leg in the same network and are both connected to the branch-office-balance-one.

Is there any possibility to debug this?


#4

@Zach_Tangen
Can you help me in any way? I just cannot solve the problem

Detailed Szenario is:

Everything is working fine.

  • If the “DR-Hub” reboots, Peplink one does reconnect fine
  • If the “Main-Hub” fails, traffic goes on through DR-Hub
  • When “Main-Hub” comes back after failure, it shows “Route-Conflict” and Peplink-One cannot send data to the HQ-networks UNTIL i reboot DR-Hub

Please help me!


#5

Out of interest have you set the costs in all directions as each setting is a one way thing.

You should have the 20 cost set on the tunnel from the balance one to the DR-hub and from the DR-hub to the balance one.


#6

@James_Webster
Yes, Route costs are set bidirectional (through InControl).


#7

@KPS
Can you private message me the serial numbers of the three devices in this configuration? I’d like to look at the configuration.


#9

@Zach_Tangen
Did you find out something about the issue?


#11

Hi KPS,

We are still investigating this issue, and have sent an update to the Peplink Distributor who logged a ticket with support. Can you share what learned routes the Fortigate has learned?

Thanks,

Erik


#12

Hi Erik!

Here the routing-info:

fg200e_HZ_1_1 (root) # get router info ospf database router adv-router 10.9.49.2

            Router Link States (Area 0.0.0.0)

LS age: 1556
Options: 0x2 (*|-|-|-|-|-|E|-)
Flags: 0x2 : ASBR
LS Type: router-LSA
Link State ID: 10.9.49.2
Advertising Router: 10.9.49.2
LS Seq Number: 8000027c
Checksum: 0x57d3
Length: 48
Number of Links: 2

Link connected to: a Transit Network
 (Link ID) Designated Router address: 10.9.49.3
 (Link Data) Router Interface address: 10.9.49.2
  Number of TOS metrics: 0
   TOS 0 Metric: 10

Link connected to: another Router (point-to-point)
 (Link ID) Neighboring Router ID: 10.10.248.1
 (Link Data) Router Interface address: 5.192.104.49
  Number of TOS metrics: 0
   TOS 0 Metric: 10

fg200e_HZ_1_1 (root) # get router info ospf database router adv-router 10.9.49.3

            Router Link States (Area 0.0.0.0)

LS age: 7
Options: 0x2 (*|-|-|-|-|-|E|-)
Flags: 0x2 : ASBR
LS Type: router-LSA
Link State ID: 10.9.49.3
Advertising Router: 10.9.49.3
LS Seq Number: 800006fd
Checksum: 0xb317
Length: 36
Number of Links: 1

Link connected to: another Router (point-to-point)
 (Link ID) Neighboring Router ID: 10.10.248.1
 (Link Data) Router Interface address: 5.192.3.177
  Number of TOS metrics: 0
   TOS 0 Metric: 110

fg200e_HZ_1_1 (root) # get router info ospf neighbor all

OSPF process 0:
Neighbor ID Pri State Dead Time Address Interface
10.9.49.2 1 Full/Backup 00:00:31 10.9.49.2 TransferPepl949
10.9.49.3 1 Full/DR 00:00:37 10.9.49.3 TransferPepl949

Can I provide something else?

Regards,
Phil


#13

Thanks Phil,

That is all we need for now.
James is in the process of replicating this in our test-lab as we speak.
Hopefully we can give you a workaround or solution by tomorrow.

Erik