RIP Hold down timer

ohassairi · ‎11-12-2012

hello

i am trying to test RIP timers

my diagram is as follow:

using default timers i config s1/0 of R1 as passive so after 3min R2 will see R1 LAN as possibly down and this route will enter the hold down state.

as i know: the hold down is defined as how long (in seconds) to “not believe” any equal or less impressive (worse) route updates for routes that are in hold down. by default HD=3min

however after 1min the route will be flushed in R2 and a new RIP route will be installed via R3!!!!. i think this should not be happen since : HD did not expire and the new route is worse.

am i right ?

Sudeep Valengattil · ‎11-13-2012

Hi,

After 240 secs, once the route is flushed there is no holddown timer associated with it, as it is removed from the RIP database. The next update which it receives from any neighbor, would be added into the routing table.

Regards,

Sudeep

ohassairi · ‎11-13-2012

thanks sudeep

so if once the route is flushed there is no holddown timer!

so why cisco choose HD to be 240s?

i think it should be just 60s so that it will expire when the route will be flushed.

i think it has no sens to choose HD so greate

blau grana · ‎11-13-2012

|------|

|------|------|------|------|

|------|------|------|------|------|

1----2-----3-----4-----5-----6

1 - update is received

2 - last update is received(hold and flush timers start to count down)

5 - hold down timer expired - routes marked as possible down

6 - flush timer expired - routes are flushed from rip database and new routes are installed in routing table if available

Best Regards Please rate all helpful posts and close solved questions

blau grana · ‎11-13-2012

i tried your scenario with default parameters and everything is working how it supose to.

Invalid after 180 seconds, hold down 180, flushed after 240

after 3mins of no updates, routes are marked in rip database as "possibly down" and after 4mins of no updates, routes are flushed from rip database. These timers start count down together, not one after another.

R3#sh clock

*00:22:24.111 UTC Fri Mar 1 2002

-

-------- R1's S0/1 made passive, R1-s0/1 <--> R3-s0/0 ---------

-

R3#sh cdp nei | i Dev|R1

Device ID Local Intrfce Holdtme Capability Platform Port ID

R1 Ser 0/0 136 R S I 3725 Ser 0/1

R3#sh ip route rip

100.0.0.0/24 is subnetted, 1 subnets

R 100.0.0.0 [120/1] via 10.0.13.1, 00:00:34, Serial0/0

R 20.0.0.0/8 [120/1] via 10.0.23.2, 00:00:20, Serial0/1

10.0.0.0/24 is subnetted, 3 subnets

R 10.0.12.0 [120/1] via 10.0.23.2, 00:00:20, Serial0/1

[120/1] via 10.0.13.1, 00:00:34, Serial0/0

R3#sh ip rip data

10.0.0.0/8 auto-summary

10.0.12.0/24

[1] via 10.0.13.1, 00:00:34, Serial0/0

[1] via 10.0.23.2, 00:00:20, Serial0/1

10.0.13.0/24 directly connected, Serial0/0

10.0.23.0/24 directly connected, Serial0/1

20.0.0.0/8 auto-summary

20.0.0.0/8

[1] via 10.0.23.2, 00:00:20, Serial0/1

30.0.0.0/8 auto-summary

30.0.0.0/8 directly connected, Loopback0

100.0.0.0/8 auto-summary

100.0.0.0/24

[1] via 10.0.13.1, 00:00:34, Serial0/0

R3#sh clock

*00:25:14.291 UTC Fri Mar 1 2002

R3#sh ip route rip

100.0.0.0/24 is subnetted, 1 subnets

R 100.0.0.0 [120/1] via 10.0.13.1, 00:02:59, Serial0/0

R 20.0.0.0/8 [120/1] via 10.0.23.2, 00:00:02, Serial0/1

10.0.0.0/24 is subnetted, 3 subnets

R 10.0.12.0 [120/1] via 10.0.23.2, 00:00:02, Serial0/1

[120/1] via 10.0.13.1, 00:02:59, Serial0/0

R3#sh ip rip data

10.0.0.0/8 auto-summary

10.0.12.0/24

[1] via 10.0.13.1, 00:02:59, Serial0/0

[1] via 10.0.23.2, 00:00:02, Serial0/1

10.0.13.0/24 directly connected, Serial0/0

10.0.23.0/24 directly connected, Serial0/1

20.0.0.0/8 auto-summary

20.0.0.0/8

[1] via 10.0.23.2, 00:00:02, Serial0/1

30.0.0.0/8 auto-summary

30.0.0.0/8 directly connected, Loopback0

100.0.0.0/8 auto-summary

100.0.0.0/24

[1] via 10.0.13.1, 00:02:59, Serial0/0

R3#

-

-------- routes via R1 marked as "possibly down" ---------

-

R3#

R3#sh clock

*00:25:19.547 UTC Fri Mar 1 2002

R3#sh ip route rip

100.0.0.0/24 is subnetted, 1 subnets

R 100.0.0.0/24 is possibly down,

routing via 10.0.13.1, Serial0/0

R 20.0.0.0/8 [120/1] via 10.0.23.2, 00:00:08, Serial0/1

10.0.0.0/24 is subnetted, 3 subnets

R 10.0.12.0 [120/1] via 10.0.23.2, 00:00:08, Serial0/1

R3#sh ip rip data

10.0.0.0/8 auto-summary

10.0.12.0/24

[1] via 10.0.23.2, 00:00:08, Serial0/1

10.0.13.0/24 directly connected, Serial0/0

10.0.23.0/24 directly connected, Serial0/1

20.0.0.0/8 auto-summary

20.0.0.0/8

[1] via 10.0.23.2, 00:00:08, Serial0/1

30.0.0.0/8 auto-summary

30.0.0.0/8 directly connected, Loopback0

100.0.0.0/8 is possibly down

100.0.0.0/24 is possibly down

R3#sh clock

*00:26:17.163 UTC Fri Mar 1 2002

R3#sh ip route rip

100.0.0.0/24 is subnetted, 1 subnets

R 100.0.0.0/24 is possibly down,

routing via 10.0.13.1, Serial0/0

R 20.0.0.0/8 [120/1] via 10.0.23.2, 00:00:06, Serial0/1

10.0.0.0/24 is subnetted, 3 subnets

R 10.0.12.0 [120/1] via 10.0.23.2, 00:00:06, Serial0/1

R3#sh ip rip data

10.0.0.0/8 auto-summary

10.0.12.0/24

[1] via 10.0.23.2, 00:00:06, Serial0/1

10.0.13.0/24 directly connected, Serial0/0

10.0.23.0/24 directly connected, Serial0/1

20.0.0.0/8 auto-summary

20.0.0.0/8

[1] via 10.0.23.2, 00:00:06, Serial0/1

30.0.0.0/8 auto-summary

30.0.0.0/8 directly connected, Loopback0

100.0.0.0/8 is possibly down

100.0.0.0/24 is possibly down

R3#

-

-------- routes via R1 flushed ---------

-

R3#

R3#sh clock

*00:26:21.359 UTC Fri Mar 1 2002

R3#sh ip route rip

R 20.0.0.0/8 [120/1] via 10.0.23.2, 00:00:11, Serial0/1

10.0.0.0/24 is subnetted, 3 subnets

R 10.0.12.0 [120/1] via 10.0.23.2, 00:00:11, Serial0/1

R3#sh ip rip data

10.0.0.0/8 auto-summary

10.0.12.0/24

[1] via 10.0.23.2, 00:00:11, Serial0/1

10.0.13.0/24 directly connected, Serial0/0

10.0.23.0/24 directly connected, Serial0/1

20.0.0.0/8 auto-summary

20.0.0.0/8

[1] via 10.0.23.2, 00:00:11, Serial0/1

30.0.0.0/8 auto-summary

30.0.0.0/8 directly connected, Loopback0

R3#sh ip proto

Routing Protocol is "rip"

Outgoing update filter list for all interfaces is not set

Incoming update filter list for all interfaces is not set

Sending updates every 30 seconds, next due in 22 seconds

Invalid after 180 seconds, hold down 180, flushed after 240

Best Regards Please rate all helpful posts and close solved questions

Sudeep Valengattil · ‎11-13-2012

Hi,

I dont agree with that.

The holddown timer would start only after the route get invalid (or possibly down state). You can confirm this with a debug. After 180 secs it would send a message, its entering holddown state.

Referring to the network diagram mentioned on top, take the case if R1 stops sending updates to R2;

1. Till 180 secs the route would be valid on R2

2. The update coming from R3 during the same time, would have a higher metric and would not be installed. This is because we have a better metric valid route already received and installed directly from R1.

3. Now Route from R1 goes invalid after 180 secs.

4. Hold down timer kicks in (meaning it will not install any other routes (say R3) till HD expires.

5. The route continues to stay in invalid (possibly down) state.

6. After 60 secs, it get flushed and removed from the RIP database and routing table.

7. Next update it receives from R3, would be added into routing table and RIP database.

regards,

Sudeep