OPERATIONAL DEFECT DATABASE
...
...
Single BGP neighbor keeps flapping. After 5 or more flaps a full reset of the BGP NSR Finite State Machine (FSM) causes BGP to change state to Not NSR-Ready. BGP instability and NSR-Not-Ready status may remain for up to 50 min. Single neighbor keeps flapping Sep 17 18:51:05 MET: bgp[1058]: %ROUTING-BGP-5-IO_INITSYNC_FAILED : No such file or directory - set id 1 Sep 17 18:51:05 MET: bgp[1058]: %ROUTING-BGP-5-IO_INITSYNC_FAILED : No such file or directory - set id 1 Sep 17 18:51:05 MET: bgp[1058]: %ROUTING-BGP-5-ADJCHANGE : neighbor 172.21.50.182 Down - BGP Notification sent, hold time expired (VRF: prov_mi_cpi) (AS: 65497) Sep 17 18:50:20 MET: bgp[1058]: %ROUTING-BGP-5-ADJCHANGE : neighbor 172.21.50.182 Up (VRF: prov_mi_cpi) (AS: 65497) Sep 17 18:49:16 MET: bgp[1058]: %ROUTING-BGP-5-IO_INITSYNC_FAILED : No such file or directory - set id 1 Sep 17 18:49:16 MET: bgp[1058]: %ROUTING-BGP-5-IO_INITSYNC_FAILED : No such file or directory - set id 1 Sep 17 18:49:16 MET: bgp[1058]: %ROUTING-BGP-5-ADJCHANGE : neighbor 172.21.50.182 Down - BGP Notification sent, hold time expired (VRF: prov_mi_cpi) (AS: 65497) Sep 17 18:48:31 MET: bgp[1058]: %ROUTING-BGP-5-ADJCHANGE : neighbor 172.21.50.182 Up (VRF: prov_mi_cpi) (AS: 65497) Sep 17 18:45:13 MET: bgp[1058]: %ROUTING-BGP-5-IO_INITSYNC_FAILED : No such file or directory - set id 1 Sep 17 18:45:13 MET: bgp[1058]: %ROUTING-BGP-5-IO_INITSYNC_FAILED : No such file or directory - set id 1 Sep 17 18:45:13 MET: bgp[1058]: %ROUTING-BGP-5-ADJCHANGE : neighbor 172.21.50.182 Down - BGP Notification sent, hold time expired (VRF: prov_mi_cpi) (AS: 65497) Sep 17 18:44:28 MET: bgp[1058]: %ROUTING-BGP-5-ADJCHANGE : neighbor 172.21.50.182 Up (VRF: prov_mi_cpi) (AS: 65497) Sep 17 18:43:23 MET: bgp[1058]: %ROUTING-BGP-5-IO_INITSYNC_FAILED : No such file or directory - set id 1 Sep 17 18:43:23 MET: bgp[1058]: %ROUTING-BGP-5-IO_INITSYNC_FAILED : No such file or directory - set id 1 Full BGP NSR FSM reset Sep 17 18:51:15 MET: bgp[1058]: %ROUTING-BGP-5-NSR_STATE_CHANGE : Changed state to Not NSR-Ready The condition of full BGP NSR FSM reset can trigger ddts "upon full reset of BGP NSR FSM, TCP returns less number of sessions in session set"; this can lead to BGP instability and NSR-Not-Ready status for up to 50 min
Continuous flap of PE-CPE EBGP sessions. One single unstable session is enough to trigger the issue
[option 1] Stop neighbor flap and avoid reaching 5 tcp init sync reset [option 2] disable NSR what is the recovery once BGP is in Not NSR-Ready? in repro environment, after about 20 min we go back into NSR-Ready. In prod environment, we experienced up to 50 min After 50 min, we recovered shutting down the EBGP neighborship
Frequent flaps of a single PE-CPE EBGP session may lead to a full reset of the BGP NSR Finite State Machine (FMS). This happens because TCP init sync fails for 5 times as the BGP neighbour keeps flapping. This fix rolls back the overly-protecting full reset for the reason TCP Init Sync fails for 5 times in a row. Meaning that the fix which has been provided it is just to avoid the full reset if TCP Init Sync fails for 5 times in a row, even from a single session
Click on a version to see all relevant bugs
Cisco Integration
Learn more about where this data comes from
Bug Scrub Advisor
Streamline upgrades with automated vendor bug scrubs
BugZero Enterprise
Wish you caught this bug sooner? Get proactive today.
