Loading...
Loading...
BFD fails to propagate sessions state change during blade restart.
The BFD session remains in the BFD sessions table and remains there until BGP session is timed out by hold the timer (90 seconds, by default). Dynamic routes, which are learnt via affected BGP session, remain in the routing table until the hold time is reached.
-- On a chassis with multiple blades, several routing protocol sessions are established, (e.g., BGP sessions). -- BFD sessions are configured for each BGP session to sustain fast failover of BGP sessions. -- There is a BGP session that can be established only via specific blade and the corresponding BFD session of this BGP session is processed on the same blade. -- This blade is restarted (e.g., using the bladectl command) or experienced a blade failure.
Change BGP hold time to reasonable lower value.
The affected BFD session is removed from the BFD table after blade reset during the period configured for this BFD session.
Click on a version to see all relevant bugs
F5 Integration
Learn more about where this data comes from
BugZero Plan
Streamline upgrades with automated vendor bug scrubs
BugZero Prevent
Wish you caught this bug sooner? Get proactive today.