I have a customer who cannot access gamepedia dot com. The ip of the server its hosted on is 141.101.114.101
When I log into my BGP router which the customers traffic is going out through I get this ping pong effect.
10.12.130.1 is the BGP router
10.12.130.254 is an internal ospf router
If I do an IP ROUTE PRINT on the subnet 141.101.114.0/24 from the BGP router, I get this
[admin@BGP_ROUTER] > ip route print detail where dst-address=141.101.114.0/24
Flags: X - disabled, A - active, D - dynamic, C - connect, S - static, r - rip, b - bgp, o - ospf, m - mme,
B - blackhole, U - unreachable, P - prohibit
0 ADb dst-address=141.101.114.0/24 gateway=10.255.255.1
gateway-status=10.255.255.1 recursive via 10.12.130.254 ETHER_2 distance=200 scope=40 target-scope=30
bgp-as-path="34245,13335,13335,13335" bgp-local-pref=100 bgp-origin=igp
received-from=BGP1-BGP2 (IBGP)
[admin@BGP_ROUTER] >
Where 10.255.255.1 is my 2nd BGP router.
It looks as though the BGP router is trying to reach the 141.101.114.101 address via the 2nd BGP router as it must think this is the best route, but this has this website broke on me now. I have loads of other sites also that users can not access, and I believe if I can fix this problem, the rest of my issues will go away also.
Anyone with any insight to a solution or even a reason why this is doing this?
Ok I know which router is sending the route back to the internal router, it is the BGP router (10.12.130.1) that the customers traffic is going out and coming in from. But why I do not know?
The customers traffic should not be going anywhere near the other BGP router unless there is a failure with the 1st BGP router.
What evidence am I looking for here in the routing table?
Another strange thing I just noticed too. If I go into routing/bgp/advertisements and do a prefix search, nothing comes up of any prefix’s I search, only my own advertised prefix’s and the prefix’s that I am having trouble with.
I understood that lets say I done a search on googles dns ip, 8.8.8.0/24 it would show me some results. Is this normal too?
I suspect that at some point certain networks had static entries added and you are now seeing confusion caused by those entries being passed in iBGP. Check the second BGP router for any routing entries involving the affected networks.
When I do a traceroute on the 2nd BGP router to the affected networks, the trace is fine, but when I do the same traces on the 1st BGP router, they just ping pong from the affected BGP router to the next internal router on the way into the network and back again.
By second BGP router are you referring to the router at 10.12.130.254 which is also running OSPF?
Did you set this system up or did somebody else? I suspect that somewhere you have entries for those “problem” networks and they are creating a routing loop. You should be able to determine the sources by looking at the routing tables and other routing entries (e.g. policy rules) on each router.
