03-24-2013 11:12 PM - edited 03-04-2019 07:23 PM
Hi guys,
We have two router on our network, each router is connected to two different ISP. When I perform an auto-fail oever test (If ISP A is down on router A, internet connection should not be affected because router B with ISP B is still connected and vice versa), the first router worked perfectly. However, on the second router, it takes 3-5 minutes to be able to go online again with ISP A. So there was downtime...
Could anyone please help me on this?
Here's the config:
Router A:
interface GigabitEthernet0/1
description upstream to ISP A
ip address 210.xxx.xxx.xxx 255.255.255.252
load-interval 30
duplex full
speed 100
media-type rj45
Router B:
interface GigabitEthernet0/1
description upstream to ISP B
ip address 125.xxx.xxx.xxx 255.255.255.252
duplex auto
speed auto
media-type rj45
Could this load-interval be the reason?
I would appreciate your soonest response.
Regards,
Jenalyn Forbes
03-25-2013 02:56 AM
Hello,
Could you please describe your connection little bit further, connection between routers, routing protocols, configuration...
load-interval is definitely not the cause of your problems ->
https://supportforums.cisco.com/thread/149119
Best Regards
Please rate all helpful posts and close solved questions
03-25-2013 07:47 PM
03-26-2013 01:33 AM
Hi,
Issue might be because you have wong iBGP peering between routerA/routerB.
RouterA:
interface Loopback0
ip address 116.1.7.1 255.255.255.255
router bgp 45941
no bgp fast-external-fallover
bgp log-neighbor-changes
neighbor iBGP peer-group
neighbor iBGP update-source Loopback0
neighbor iBGP version 4
neighbor eBGP peer-group
neighbor eBGP version 4
neighbor 116.1.6.1 remote-as 12345
neighbor 116.1.6.1 peer-group iBGP
neighbor 210.1.2.3 remote-as 6789
neighbor 210.1.2.3 peer-group eBGP
RouterB:
interface Loopback0
ip address 116.1.6.1 255.255.255.255
router bgp 45941
no bgp fast-external-fallover
bgp log-neighbor-changes
neighbor iBGP peer-group
neighbor iBGP update-source Loopback0
neighbor iBGP version 4
neighbor eBGP peer-group
neighbor 116.1.7.1 remote-as 45941
neighbor 116.1.7.1 peer-group iBGP
neighbor 125.1.2.3 remote-as 6648
neighbor 125.1.2.3 peer-group eBGP
Does second router have a default route?
Hope it will help.
Best regards,
Abzal
03-26-2013 01:42 AM
Hi Abzal,
Sorry, it's a typo error when I edit the config for you.
It is actually:
Router A:
router bgp 12345
no bgp fast-external-fallover
bgp log-neighbor-changes
neighbor iBGP peer-group
neighbor iBGP update-source Loopback0
neighbor iBGP version 4
neighbor eBGP peer-group
neighbor eBGP version 4
neighbor 116.1.6.1 remote-as 12345
neighbor 116.1.6.1 peer-group iBGP
neighbor 210.1.2.3 remote-as 6789
neighbor 210.1.2.3 peer-group eBGP
Router B:
router bgp 12345
no bgp fast-external-fallover
bgp log-neighbor-changes
neighbor iBGP peer-group
neighbor iBGP update-source Loopback0
neighbor iBGP version 4
neighbor eBGP peer-group
neighbor 116.1.7.1 remote-as 12345
neighbor 116.1.7.1 peer-group iBGP
neighbor 125.1.2.3 remote-as 6648
neighbor 125.1.2.3 peer-group eBGP
Sorry for the confusion.
Please check where else can be the possible cause of this issue.
Regards,
Jenalyn
03-26-2013 01:50 AM
Hi Jenalyn,
Ok, I see.
Then question: do you have default routes from perspective ISPs on both routers? If yes, then I suppose router/switch that connected to both routers(routerA/routerB) might be doing load-balancing between them. You may check it.
And try to change OSPF configuration on routerA/routerB:
router ospf 1
no default-information originate always
default-information originate
Because with always keyword router will be advertising default route to other OSPF routers even it doesn't have one.
http://www.cisco.com/en/US/tech/tk365/technologies_configuration_example09186a00801ec9f0.shtml
Hope it will help.
Best regards,
Abzal
03-26-2013 02:20 AM
Also for iBGP neighbors you should have configured next-hop-self ->
router bgp 12345
neighbor iBGP peer-group next-hop-self
Beause iBGP peers do not change next hop IPs when advertising prefixes to iBGP peers by default. This could be problem, RouterA will not use prefixes via RouterB and vice versa because next hop is unreachable for them.
You should either configure next-hop-self for iBGP peer-group or use OSPF to advertise next-hops -> include WAN interfaces in OSPF but make them passive.
Best Regards
Please rate all helpful posts and close solved questions
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide