08-02-2007 10:43 AM - edited 03-03-2019 06:09 PM
Hi!
I have 3 E1 circuits connected from 7613 Router to 3640. I?m trying to configure load-balance to share these circuits (by OSPF), but, I have no success. Look the sh ip route and ospf below:
sh ip route vrf vrf_name ospf
O E2 1.1.1.1 [110/20] via 1.1.116.50, 4d06h, Serial2/1/3:1
[110/20] via 1.1.116.46, 4d06h, Serial2/1/6:1
[110/20] via 1.1.116.62, 4d06h, Serial2/1/4:1
router#sh ip ospf interface bri
Interface PID Area IP Address/Mask Cost State Nbrs F/C
Se2/1/6:1 190 2 1.1.116.45/30 50 P2P 1/1
Se2/1/3:1 190 2 1.1.116.49/30 50 P2P 1/1
Se2/1/4:1 190 2 1.1.116.61/30 50 P2P 1/1
Same cost and balancing work, correct? NO! I know that load-balancing per-packet not work at 76XX routers and MlPPP is not a option here. What can I do to share per equal cost balance here?
08-02-2007 12:04 PM
I believe you could accomplish per-packet-load balancing through cef.
Cisco Express Forwarding config
CEF Overview and Restricted platforms
08-02-2007 12:21 PM
Thanks Jorge!
This solution is possible only on 3640 (in this case), but on 7613 is not possible!
Per-packet feature not appear available on 76XX routers.
Anybody have any idea?
08-02-2007 12:37 PM
Leandro
I do not understand what the problem is. Your original post shows that the IP routing table has 3 equal paths (at least to 1.1.1.1) and with 3 equal paths the packet forwarding should share the load over the links. Can you demonstrate that load sharing is not happening? I would suggest this demonstration:
- show interface for the 3 serial interfaces and note the packet sent counter for each interface.
- use extended ping to ping 1.1.1.1 and send 3000 ping packets.
- show interface for the 3 serial interface and note the packet sent counter for each interface.
HTH
Rick
08-03-2007 06:15 AM
08-03-2007 08:29 AM
Leandro
I have looked at the output that you posted and to my perspective it does demonstrate that load sharing is being done - there are output packets on all 3 interfaces. The load is not evenly divided, but it is divided. This is all that OSPF can accomplish and there is not a problem with OSPF. The result that you are seeing is sometimes caused when there is a destination that is receiving a larger amount of traffic than other destinations. If you want to investigate this you might turn on IP accounting or turn on NetFlow to look at traffic and report sources and destinations.
HTH
Rick
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: