Query regarding track command

[mukundh86]

Hi

Can someone give me an explanation of what the "delay" command exactly does under the "track" command and how does ot affect the static routes. I have stated the configurations as follows

track 100 rtr 100 reachability
delay down 90 up 120

ip route 0.0.0.0 0.0.0.0 serial0/0 10 track 100
ip route 0.0.0.0 0.0.0.0 Dialer1 251

The object id 100 was configured with the "ip sla monitor" command and that part was clear as it was mainly there to track serial0/0 and check if it goes down or not and if it does the default route should point to the ADSL-Dialer interface. Only these above listed configs I am not able to understand clearly especially the "delay" part.

[Richard Burts]

Mukundh

In this configuration the delay parameter is used to adjust for the possibility of a flapping interface. The track is testing reachability through some interface. So the question becomes: if the reachability becomes unreachable do you immediately want to remove the original route and replace it with the alternate (higher AD) route. The complication is what to do in a situation where the interface goes down, and quickly, comes back up. If it has quickly come back up, do you really want to change routes and then change them back again. By specifying delay you are saying: if it goes down and stays down for x then remove the original route and replace it with the alternate route.

HTH

Rick

[mukundh86]

Hi Richard

Thanks a lot for the explanation. So as per the command the router will take the alternate route if the first route is down for 90 seconds and come back to thw original route if it is up for 120 seconds right? Is there any other significance of the ADs apart from AD of one route being lower than the other?

Thanks

Mukundh

[Richard Burts · Accepted solution]

Mukundh

I believe that your understanding is correct. The router will wait 90 seconds after detecting the tracked object down to make sure that it is not a quick/transient outage that will come back within 90 seconds. And after it has been down and is detected as back up it will wait 120 to make sure that it stays up (in many outage situations the link may come back up briefly but then go down again - so this delay in using it again is to make sure that it is stable).

The first static route does not specify an AD and so gets the default AD for static routes of 1.

ip route 0.0.0.0 0.0.0.0 serial0/0 10 track 100

The second static route specifies an AD of 251. The main significance of the AD is that the second one is higher than the first one. So in normal operation only the first one with lower AD will be used and the second one is not used. If the first one becomes down (due to track) then the second one with AD 251 will be used. (This is frequently referred to as a floating static route)

There is perhaps some additional significance in the choice of 251 as the AD for the second route. This AD makes the route less desirable than the normal static route and also less desirable than a route that was advertised by one of the dynamic routing protocols (such as OSPF or EIGRP).

So let us think of a scenario that shows how this works. In this scenario we have the first static route (with default AD of 1), perhaps we have an OSPF route, and the second static (with AD of 251). The first static has the most favorable AD so it is used. The AD of OSPF is between the AD of 1 and of 251, so if the first static is withdrawn (due to track) then the OSPF route would be used. If the first static is down and something happens to the OSPF route that causes it to be withdrawn, then the second static (with AD 251) would be used.

HTH

Rick

[mukundh86]

Thanks a lot Rick for the detailed explanation. It has given me a clear idea of the track command and it's usage. Thanks a million.