Routed and SVI failovers

In some situations with various technologies you require ultra fast fail-over. There are design considerations in which you may require a lower fail-over than 200 millisecond. Lets look at how you can improve this. This post looks at L2 switch ports with a L3 Switched interface versus a L3 routed port.

Layer 2 Switch port and Layer 3 Switched Virtual Interface

1. Link Down
2. Interface Down
3. Autostate
4. SVI Down
5. Routing Update

14:12:47.312 AEST: %LINEPROTO-5-UPDOWN: Line protocol
on Interface GigabitEthernet6/1, changed state to down
14:12:47.320 AEST: %LINK-3-UPDOWN: Interface
GigabitEthernet6/1, changed state to down
14:12:48.568 AEST: %LINK-3-UPDOWN: Interface Vlan3,
changed state to down
14:12:48.568 AEST: IP-EIGRP(Default-IP-Routing-Table:100):
Callback: route, adjust Vlan3

The fail-over of 150-200 msec is quite impressive. You can see that the protocol drops, interface goes down, the SVI fails then the route is withdrawn via EIGRP. It is possible though to shave this down to around 10 msec with a routed interface.

Layer 3 Routed interface

The order of operations for a Layer 3 routed interface doesn’t have auto-state calculations or have the need to wait on SVI state up and down. Lets see the following.

1. Link Down
2. Interface Down
3. Routing Update

13:16:37.645 AEST: %LINEPROTO-5-UPDOWN: Line
protocol on Interface GigabitEthernet3/1, changed state
to down
13:16:37.653 AEST: %LINK-3-UPDOWN: Interface
GigabitEthernet3/1, changed state to down
13:16:37.653 AEST: IP-EIGRP(Default-IP-Routing-
Table:100): Callback: route_adjust GigabitEthernet3/1

8 msec loss or there about. There are two extra steps we have culled from the fail over process. It is worth considering in a highly time sensitive environment. These are just some musings inspired by Cisco Live 365 presentation Next Generation Campus Architectures BRKCRS-2663 and is available for viewing at Ciscolive365