You want to adjust RIP timers on a ScreenOS firewall to be different from the default values.
To achieve faster convergence, you can adjust the invalid and flush timers in the ScreenOS RIP implementation downward:
set vrouter trust-vr protocol rip invalid-timer 90extfw->
set vrouter trust-vr protocol rip flush-timer 60
Thus, the duration for a stale route to be removed from a ScreenOS firewall's routing table is reduced to as low as 180 seconds from the default value of at least 300 seconds. Note that the update and hold-down timers are kept at their default values.
Invalid, flush, and hold timers are implemented in most modern RIP implementations to prevent routing loops and flapping routes. These timers are relevant in scenarios where there is a "silent" failure—for example, a RIP neighbor that was advertising the routes goes down, but the local physical interface on which the routes were being received is still up. If a local physical interface on which RIP routes are being learned goes down, the associated routes are immediately invalidated and purged from the routing table.
The default values for the various RIP timers relevant to convergence, excluding demand-circuit environments, in ScreenOS are as follows:
Defines the frequency of RIP
RESPONSE messages with route entries: every 30 seconds.
Defines the duration that a ScreenOS device waits before marking ...