Bonding links between 1 Cisco routers and two ISP

[tefomatlho]

Hello,

I want to aggregate 2 ISP link to have one big link and thus simulate a very high bandwidth connection. I have 20Mbps on the primary and 10Mbps on the secondary link.

[balaji.bandi]

There is an example and solution below link give you an idea how you an achieve this.

https://www.cisco.com/c/en/us/support/docs/ip/network-address-translation-nat/100658-ios-nat-load-balancing-2isp.html

BB

[Georg Pauwen]

Hello,

you can also use static routes. Since one of your ISPs has a 20Mbps and the other a 10Mbps, you want unequal load splitting. Let's say these are your interfaces:

interface FastEthernet0/0

description 20MBPS_LINK_TO_ISP_1

ip address 10.10.10.1 255.255.255.252

!

interface FastEthernet0/1

description 10MBPS_LINK_TO_ISP_2

ip address 20.20.20.1 255.255.255.252

The below three static routes would split traffic at a 2:1 ratio:

ip route 0.0.0.0 0.0.0.0 FastEthernet0/0

ip route 0.0.0.0 0.0.0.0 10.10.10.2

ip route 0.0.0.0 0.0.0.0 FastEthernet0/1

[tefomatlho]

HI 

i have this configs on the router, also the nat translation shows the ratio but the user machine gets 10Mbps or yes.

ip route 0.0.0.0 0.0.0.0 168.167.50.109
ip route 0.0.0.0 0.0.0.0 GigabitEthernet0/0.1954
ip route 0.0.0.0 0.0.0.0 Dialer1

[Georg Pauwen]

Hello,

you can only do per destination load balancing. Post the full config of your router so we can see the NAT setup...

[a.alekseev]

Int isp1
Desc 10mbps
Ip add 10.0.0.0 255.255.255.254

Int isp2
Desc 20mbps
Ip add 20.0.0.0 255.255.255.254

For ratio 2:1
Let's make a hack. Place a fake route. Let it be 5.5.5.0/31

Ip route 5.5.5.5 255.255.255.255 20.0.0.1

Ip route 0.0.0.0 0.0.0.0 5.5.5.0
Ip route 0.0.0.0 0.0.0.0 5.5.5.1
Ip route 0.0.0.0 0.0.0.0 10.0.0.1

[tefomatlho]

Does thi mean i will have to change my routes, remember this is a production network.

Also i think the below statement address the fake route //Correct me

ip route 0.0.0.0 0.0.0.0 168.167.50.109
ip route 0.0.0.0 0.0.0.0 GigabitEthernet0/0.1954

ip route 0.0.0.0 0.0.0.0 Dialer1

[a.alekseev]

sh ip cef 0.0.0.0/0 internal

[tefomatlho]

Hi

i am not sure if the below will help, our redundant link is still down the ISP is on it.

HAN-C1921-SER#sh ip cef 0.0.0.0/0 in
HAN-C1921-SER#sh ip cef 0.0.0.0/0 internal
0.0.0.0/0, epoch 0, flags [att, cover, defrt], RIB[S], refcnt 5, per-destination sharing
sources: RIB, DRH
feature space:
IPRM: 0x00048004
subblocks:
Covered dependent prefixes: 14417
notify cover updated: 14417
ifnums:
GigabitEthernet0/0.1954(11): 168.167.50.109
Dialer1(13)
path list 2913501C, 3 locks, per-destination, flags 0x69 [shble, rif, rcrsv, hwcn]
path 3338CD58, share 1/1, type attached prefix, for IPv4
attached to GigabitEthernet0/0.1954, glean
path 33387C18, share 0/1, type attached prefix, for IPv4
attached to Dialer1, IP adj out of Dialer1 (incomplete)
path 33393B70, share 1/1, type recursive, for IPv4
recursive via 168.167.50.109[IPv4:Default], fib 2D7EEA04, 1 terminal fib, v4:Default:168.167.50.109/32
path list 2BBAB7C8, 3 locks, per-destination, flags 0x4A [nonsh, rif, hwcn]
path 2BBAA828, share 1/1, type adjacency prefix, for IPv4
attached to GigabitEthernet0/0.1954, IP adj out of GigabitEthernet0/0.1954, addr 168.167.50.109 2D7EB7A0
output chain:
loadinfo 31159438, per-session, 3 choices, flags 0003, 5 locks
flags [Per-session, for-rx-IPv4]
15 hash buckets
< 0 > glean
< 1 > IP adj out of Dialer1 (incomplete)
< 2 > IP adj out of GigabitEthernet0/0.1954, addr 168.167.50.109 2D7EB7A0
< 3 > glean
< 4 > IP adj out of Dialer1 (incomplete)
< 5 > IP adj out of GigabitEthernet0/0.1954, addr 168.167.50.109 2D7EB7A0
< 6 > glean
< 7 > IP adj out of Dialer1 (incomplete)
< 8 > IP adj out of GigabitEthernet0/0.1954, addr 168.167.50.109 2D7EB7A0
< 9 > glean
<10 > IP adj out of Dialer1 (incomplete)
<11 > IP adj out of GigabitEthernet0/0.1954, addr 168.167.50.109 2D7EB7A0
<12 > glean
<13 > IP adj out of Dialer1 (incomplete)
<14 > IP adj out of GigabitEthernet0/0.1954, addr 168.167.50.109 2D7EB7A0
Subblocks:
None
HAN-C1921-SER#

[Joseph W. Doherty]

Perhaps the "best" way to take advantage of your two (unequal bandwidth) link would be to use PfR (if supported on your platform).

PfR can dynamically load balance your links, including taking into account their different bandwidth capacities.

[tefomatlho]

Find the below out puts;

HAN-C1921-SER#SH DOMain MACKAIR BOrder ALL

Thu Jul 05 18:46:27.996
-------------------------------------------------------------------------------------Borders Configured:

VRF: default
Instance Status: UP
Present status last updated: 00:04:45 ago
Loopback: Configured Loopback0 UP (10.8.1.1)
-------------------------------------------------------------------------------------

HAN-C1921-SER#SH DOMain MACKAIR BOrder STATU

Thu Jul 05 18:48:56.476
-------------------------------------------------------------------------------------
****Border Status****

Instance Status: UP
Present status last updated: 00:02:10 ago
Loopback: Configured Loopback0 UP (10.8.1.1)
Master: 10.10.4.1
Connection Status with Master: DOWN (retry in 00:00:00)
MC connection info: Socket Pending: Socket is not connected
Disconnected for: 00:02:10
Route-Control: Enabled
Minimum Mask length: 28
Sampling: off
Minimum Requirement: Not Met
PMI update: Not received
Globals Update: Not received
(Will attempt shut/no-shut if min requirement not meet in 169 secs)
External Wan interfaces:

Auto Tunnel information:

Name:Tunnel0 if_index: 20
Borders reachable via this tunnel:
-------------------------------------------------------------------------------------

[Joseph W. Doherty]

Was there a question meant to be added to this post?

[tefomatlho]

Hi

i want to know is i can achieved bonded bandwidth using 2 ISP's 

[a.alekseev]

All options have been said.

If you want to have 30Mbps per one flow - it's not possible.

If you need load-balancing - you can try pfr

https://www.cisco.com/c/en/us/products/ios-nx-os-software/performance-routing-pfr/index.html

or

just simple as a nail

unequal load balancing with static routes.