No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

Half of RNC Traffic dropped by PE NE40E becasue of bad static route design on LSW S9306

Publication Date:  2012-07-27 Views:  38 Downloads:  0
Issue Description
The customer Network is NGN which all the CS ,PS and Acess network are interconnected via IP BB.
The CS contains MSS,MGW and GMSS .
The PS Ccontains SGSN and GGSN.
The IPBB contrains NE40E V600R001C00SPC800 as PEs and NE40E V600R001C00SPC800 as Ps.
The Acess network contains RNCs which are connected to S9306 SW then to PE
So the traffic will go from RNC then to LSW S9306 then via static routes it will load sharing by both PE01 and PE02 and then to MGW and then to MSS.
in the normal situation every think is ok but when the PE02 is down 50 % of the traffic dropped at the time we expext that it will be backuped by PE01



Alarm Information
NOC Team notice a lot of dropped packets by U2000
Handling Process
1-On the LSW we have two statis routes for MGW as below :

ip route-static 10.13.96.0 255.255.255.240 10.13.64.33 description MGW-PE01 // to MGW via PE01
ip route-static 10.13.96.16 255.255.255.240 10.13.64.33 description MGW-PE02 // to MGW via PE01-PE02
2-when we ping both routes we found that the the ping to MGW via PE02 is dropped
3-We fixed this issue via using BFD from LSW to PE02 and then Associate the Static roues to the BFD
ip route-static 10.13.96.16 255.255.255.240 10.13.64.33 track bfd-session LSW-to-PE02
so right now when PE02 connection is down the BFD will be down and the static route to MGW-PE02 wil be down all the traffic will be routed by PE01 via MGW-PE01 static route.


Root Cause
Becasue we use static routes from LSW to PE01 and PE02 ,so the Static routes status didnt change when PE02 went down,but both routes to PE01 and PE02 on the LSW are still up since the next hop of both routes (the physical connection from LSW to PE01) is still up,
This means that the LSW doent know that the PE02 is down so the traffic now goes from RNC to LSW then to PE01 then PE01 should forward part of the traffic direclty to MGW and the other part Part to PE02 which will be sent to MGW but this part will be dropped now since PE02 connection is down
Suggestions
So Dynamic routes is always preferd than the static routes since all the nodes in the network will be updated via the dynamic routing protocol about any topology change but in case we use static routes we should take care about this point and we can use BFD and make the static route track the BFD status to be awareness of the topology change

END