BGP flaps after OSPF flaps because of BFD

Naveed N · ‎06-20-2014

Hi

We have a setup CE Router (ASR1002) à L2 Swtich (3750) à Firewall (ASA 5540) à Core Switch (VSS 6509 – SUP2T). When OSPF falps because of BFD the BGP as well flaps extactly after some time. Is there any bug relating to this issue.

Jun 20 03:03:34 gmt: %OSPF-SW1-5-ADJCHG: Process 1, Nbr 1.1.1.3 on GigabitEthernet2/3/3 from FULL to DOWN, Neighbor Down: BFD node down

Jun 20 03:03:35 gmt: %OSPF-SW1-5-ADJCHG: Process 1, Nbr 1.1.1.3 on GigabitEthernet2/3/3 from DOWN to INIT, Received Hello

Jun 20 03:03:41 gmt: %OSPF-SW1-5-ADJCHG: Process 1, Nbr 1.1.1.3 on GigabitEthernet2/3/3 from INIT to 2WAY, 2-Way Received

Jun 20 03:03:41 gmt: %OSPF-SW1-5-ADJCHG: Process 1, Nbr 1.1.1.3 on GigabitEthernet2/3/3 from 2WAY to EXSTART, AdjOK?

Jun 20 03:03:41 gmt: %OSPF-SW1-5-ADJCHG: Process 1, Nbr 1.1.1.3 on GigabitEthernet2/3/3 from EXSTART to EXCHANGE, Negotiation Done

Jun 20 03:03:42 gmt: %OSPF-SW1-5-ADJCHG: Process 1, Nbr 1.1.1.3 on GigabitEthernet2/3/3 from EXCHANGE to LOADING, Exchange Done

Jun 20 03:03:42 gmt: %OSPF-SW1-5-ADJCHG: Process 1, Nbr 1.1.1.3 on GigabitEthernet2/3/3 from LOADING to FULL, Loading Done

Jun 20 03:05:54 gmt: %OSPF-SW1-5-ADJCHG: Process 1, Nbr 1.1.1.3 on GigabitEthernet2/3/3 from FULL to DOWN, Neighbor Down: BFD node down

Jun 20 09:02:31 gmt: %BGP-SW1-5-NBR_RESET: Neighbor 10.1.1.2 reset (Peer closed the session)

Jun 20 09:02:31 gmt: %BGP-SW1-3-NOTIFICATION: received from neighbor 10.1.1.2 4/0 (hold time expired) 0 bytes

Jun 20 09:02:31 gmt: %BGP-SW1-5-ADJCHANGE: neighbor 10.1.1.2 Down Peer closed the session

Jun 20 09:02:31 gmt: %BGP_SESSION-SW1-5-ADJCHANGE: neighbor 10.1.1.2 IPv4 Unicast topology base removed from session Peer closed the session

Jun 20 09:02:43 gmt: %BGP-SW1-5-ADJCHANGE: neighbor 10.1.1.2 Up

mhnedirli · ‎06-20-2014

hi,

this is normal behaviour, bgp tracks the next hop reachability. try to configure your OSPF neighborship. could you can send device configuration.

Naveed N · ‎06-20-2014

Hi,

Thanks for your reply, there is a gap of 6 hours, many times OSPF flaps because of BFD but BGP does not flap. BGP flaps are found exactly after 6 hours of OSPF flaps and the Core switch is terminating the BGP Session

Warm regards

Naveed

mhnedirli · ‎06-20-2014

hi,

maybe there could be another problem, BGP establish a TCP connection with peer device and periodically sends keepalive messages, by default it sends keepalives in 60 seconds period and 180 seconds hold time. you can increase hold time

neighbor x.x.x.x timers 20 60

Naveed N · ‎06-20-2014

Hi,

Thanks for your reply. The issue is strange, OSPF flaps more than 5 to 6 times a day because of BFD node down. Morning 3:02 AM the OSPF flaps, exactly at 9:02 AM BGP flaps and there is no OSPF flap after 3:02AM till 9:02AM, BGP does not flap in other intervals of OSPF flap. The BGP flap has happened for more than 4 days exactly at 9:02 AM. Increasing the timers may not help, as this is happening after 6 hours.