...
Cat9K HA setup/ 16.9.x, 16.10.x, 16.6.x Routed traffic (pass through and routed by Cat9K-HA setup running 16.9.x, 16.10.x, 16.6.x) for random IPs might be broken due to incorrect rewrite info programming in FED and will be recovered by clearing ARP cache for the given IP. At broken state, we see a mismatch in Rewrite Index between the Active/standby for broken Adj/IP, we also see that RI_HDL dump shows incorrect DMAC in FED programming. Which is why, packet is routed with incorrect Destination MAC, leading to connectivity loss. In some instances, we have seen that egress VLAN rewrite info is also incorrect in hardware FED, even though software FED entry shows correct. !
Clearing ARP for broken IP recovers the connectivity issue. A reload of affected switch should clear the issue temporarily. ! Reducing ARP timeout value may not help.
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.