cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1914
Views
0
Helpful
35
Replies

OSPF Not load balancing

Brandon Baker
Level 1
Level 1

We currently have 3 ISPs, the third just being added last week, When I set it up under OSPF in our Core switch it will not load balance, Our current defrault route in the core is going to ISP2 and when I add our new connection in as a default route it stelas all of the traffic from from ISP2 instead of load balancing. Can anyone please assist?

      

All 3 have same Costs as well

35 Replies 35

Yes that is what is happening, what should I look at in the VXR?

cas-gw-7206vxr-citynet-01# show ip protocol
Routing Protocol is "ospf 1"
  Outgoing update filter list for all interfaces is not set
  Incoming update filter list for all interfaces is not set
  Router ID X.20.128.1
  It is an autonomous system boundary router
  Redistributing External Routes from,
  Number of areas in this router is 1. 1 normal 0 stub 0 nssa
  Maximum path: 4
  Routing for Networks:
    X.20.128.0 0.0.0.127 area 0.0.0.1
  Routing Information Sources:
    Gateway         Distance      Last Update
    X.16.129.2         110      4d00h
    X.20.128.2         110      13w3d
    X.20.128.22        110      1d03h
    X.20.128.253       110      1d03h
  Distance: (default is 110)

Routing Protocol is "bgp 40098"
  Outgoing update filter list for all interfaces is not set
  Incoming update filter list for all interfaces is not set
  IGP synchronization is disabled
  Automatic route summarization is disabled
  Redistributing: ospf 1 (internal)

  Unicast Aggregate Generation:
    X.231.16.0/21     summary-only
  Neighbor(s):
    Address          FiltIn FiltOut DistIn DistOut Weight RouteMap
    X.118.67.17                  3
    X.118.67.105                 1              1        AS-6536-INCOMING-VOIP
    X.20.128.22
  Maximum path: 2
  Routing Information Sources:
    Gateway         Distance      Last Update
    (this router)        200      6d10h
    X.118.67.105         20      8w0d
    X.20.128.22        200      17w0d
    X.118.67.17          20      3d22h
  Distance: external 20 internal 200 local 200

Does it seem to be balancing via BGP? I have not set the new interface up via bgp as I thought that was per device than per interface?

I am not sure yet whether it is balancing via BGP. I am curious how the balancing was done before the new connection was brought up. Perhaps the output of show ip route (focusing especially on anything relating to 0.0.0.0 or gateway of last resort..

Also what changes were made on the ASR as the new connection was brought up? I am wondering if something was done here that impacted using the other ISP.

HTH

Rick

HTH

Rick

VXR:

Gateway of last resort is X.118.67.17 to network 0.0.0.0 (ISP2)


B*   0.0.0.0/0 [20/0] via X.118.67.17, 7w0d

No changes were made to the ASR- just activating a gigabitethernet port for the Fiber connection and activating another gigabitethernet connection to the 6509.

Thanks. Could you also post these outputs from the ASR?

HTH

Rick

HTH

Rick

I apologize, I did add the following in the ASR as well:

ip route 0.0.0.0 0.0.0.0 XXX (ISP1 NEIGHBOR)

ip route 0.0.0.0 0.0.0.0 XXX (ISP3 NEIGHBOR)

This is the Ip route

Gateway of last resort is x.185.86.205 to network 0.0.0.0

S*    0.0.0.0/0 [1/0] via x.185.86.205 (isp3)
                [1/0] via x.126.51.97 (isp1)

There is also a full bgp routing table coming from ISP1 into the ASR

Thanks for this information. But I am still not seeing anything that is sending traffic through ISP 2.

Is there any connection between the ASR and the VXR?

Are you perhaps running IBGP between ASR and VXR?

In looking at your most recent response I see you phrase it as I did add the following in the ASR as well:

and it makes me wonder what was in place before you added these?

HTH

Rick

HTH

Rick

there is no physical connection between the two- they are both in the same Vlan in the 6509 and the new connection is not. I am unaware if we are running IBGP or not, is there a way to find out? I am not the one who set this network up just jumped in after they left recently.

I just removed the 2 default routes that I added into the ASR

now I have the following:

Gateway of last resort is X.126.51.97 to network 0.0.0.0 (isp1)

B*    0.0.0.0/0 [20/0] via X.126.51.97, 00:00:27 (isp1)

the original 2 connections going to the catalyst are in the same subnet as well if that matters. The new connection is in a different subnet since it is on the ASR with the original connection.

Probably a good way to check on the BGP is the output of show ip bgp sum.

The default route that shows up when you remove the static default route is pretty much what I would expect from running BGP with ISP 1. And I would expect that when you bring up BGP with ISP 3 that they will also advertise a default route to you. But BGP with 2 ISP may not work quite the way that you expect. It is likely that both ISP will advertise a default route and your BGP will usually choose one or the other (but usually not both at the same time). If you want to do load sharing/load balancing you frequently have ISP 1 advertise some group of routes to you in addition to its default route and you have ISP 3 advertise some group of routes to you (different from what you learned from ISP 1) in addition to its default route.

I am still wondering what was sending traffic through ISP 2? So far I have not seen anything that would explain how some of your traffic was going through that ISP? But I am wondering if BGP on the catalyst might have been how that was working. In an earlier post you gave show ip protocol from the 6509 that seems to show that the ASR and the VXR were BGP neighbors. Was the static default route that we looked at in the beginning of this discussion 

Gateway of last resort is X.X.128.1 to network 0.0.0.0

S*   0.0.0.0/0 [200/0] via x.x.128.1

               [200/0] via x.x.128.139

there for a long time or was it perhaps something you added as you worked on the third ISP connection?

HTH

Rick

HTH

Rick

Without the static route for the .139 connection on the 6509 I get the following:

Gateway of last resort is X.20.128.1 to network 0.0.0.0

S*   0.0.0.0/0 [200/0] via X.20.128.1

Should I set up the .139 connection as a BGP neighbor on the catalyst as well? Or do you think it would work better if the connection from the 6509 to the routers as OSPF instead of BGP?

core-6509#show ip bgp sum
BGP router identifier X.20.135.65, local AS number 40098
BGP table version is 35288663, main routing table version 35288663
89 network entries using 10057 bytes of memory
90 path entries using 4320 bytes of memory
75/22 BGP path/bestpath attribute entries using 7500 bytes of memory
60 BGP AS-PATH entries using 1752 bytes of memory
0 BGP route-map cache entries using 0 bytes of memory
0 BGP filter-list cache entries using 0 bytes of memory
BGP using 23629 total bytes of memory
BGP activity 9269519/9269429 prefixes, 14261513/14261423 paths, scan interval 60 secs

Neighbor        V    AS MsgRcvd MsgSent   TblVer  InQ OutQ Up/Down  State/PfxRcd
X.20.128.1    4 40098  642026  641009 35288663    0    0 6d01h          89
X.20.128.2    4 40098 11715904  641040 35288663    0    0 6d01h           1

You took away one static default using .139 and still have one static default route. I wonder what would be the result if you also took away the static default using 128.1?

I am pretty sure that you will not need to set up BGP for ISP 3 on the catalyst. Assuming that it is already running BGP with the ASR (I am assuming that 128.1 and 128.2 are probably the ASR and the VXR - is that right?) then when the ASR is running BGP with ISP 3 then its routes should be shared automatically).

Obviously I do not know much about your network, but from what I have seen so far I believe that the better way to do routing is to use OSPF for the catalyst and the other routers and switches inside your network and to use BGP between your 6509, the ASR, and the VXR.

HTH

Rick

HTH

Rick

I will try taking the static route out tomorrow early AM incase of iany issues and post the results. Yes BGP is running between the devices currently and 128.1 is the VXR and 128.2 is the ASR. The ASR is running BGP with the third ISP currently (only a default routing table) and I do have traffic gong out it, it is basically splitting the traffic that is meant for ISP1.

"Obviously I do not know much about your network, but from what I have seen so far I believe that the better way to do routing is to use OSPF for the catalyst and the other routers and switches inside your network and to use BGP between your 6509, the ASR, and the VXR."

I am confused here as it looks like it is currently using BGP between the 6509, ASR, and VXR. Are you recommending running OSPF for this instead of BGP? How would I go about switching this, if I turn BGP off on the 6509 will it automatically switch to OSPF?

I also see that if I run a show ip bgp in the 6509 that it seems to be using iBGP and my new BGP peer is not listed anywhere in there:

* i0.0.0.0          X.126.51.97             0    100      0 7018 i (ISP1)

*>i                 X.118.67.17             0    100      0 6536 i (ISP2)

there are also a bunch of others with the next hop to ISP2 but the only route for ISP1 is the default route.

      

In the ASR i have the following:

     Network          Next Hop            Metric LocPrf Weight Path

*   0.0.0.0          X.185.86.205           0             0 6461 i (ISP3)

*>                   x.126.51.97                           0 7018 i (ISP1)

                      0.0.0.0                                0 i

VXR:

   Network          Next Hop            Metric LocPrf Weight Path
*> 0.0.0.0          X.118.67.17             0             0 6536 i

Getting Started

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: