GLBP and Interface Tracking

johnnylingo · ‎02-04-2010

I'm migrating a pair of routers from HSRP to GLBP. As with the HRSP, I want to track the upstream interface, and shift traffic to the other router if the interface is down.

My understand is GLBP does this using the "weighing track" command. The default weight is 100, so if I track the interface and decrement by 100, then the local router should take no traffic if the interface is down:

hostname router-A

!

track 1 interface Serial0/0 line-protocol

!

interface FastEthernet0/0
ip address 192.168.10.1 255.255.255.0
glbp 10 ip 192.168.10.254
glbp 10 priority 255
glbp 10 weighting track 1 decrement 100
!

hostname router-B

!

track 1 interface Serial0/0 line-protocol

!

interface FastEthernet0/0
   ip address 192.168.10.2 255.255.255.0
   glbp 10 ip 192.168.10.254
   glbp 10 priority 254
   glbp 10 weighting track 1 decrement 100
!

When I bring down Serial0/0, the weight is set to 0, however there is no change in the traffic weight between the two GLBP peers.

Ganesh Hariharan · ‎02-04-2010

I'm migrating a pair of routers from HSRP to GLBP. As with the HRSP, I want to track the upstream interface, and shift traffic to the other router if the interface is down.

My understand is GLBP does this using the "weighing track" command. The default weight is 100, so if I track the interface and decrement by 100, then the local router should take no traffic if the interface is down:

hostname router-A

!

track 1 interface Serial0/0 line-protocol

!

interface FastEthernet0/0
ip address 192.168.10.1 255.255.255.0
glbp 10 ip 192.168.10.254
glbp 10 priority 255
glbp 10 weighting track 1 decrement 100
!

hostname router-B

!

track 1 interface Serial0/0 line-protocol

!

interface FastEthernet0/0
   ip address 192.168.10.2 255.255.255.0
   glbp 10 ip 192.168.10.254
   glbp 10 priority 254
   glbp 10 weighting track 1 decrement 100
!

When I bring down Serial0/0, the weight is set to 0, however there is no change in the traffic weight between the two GLBP peers.

Hi ,

The GLBP group weighting can be automatically adjusted by tracking the state of an interface within the router. If a tracked interface goes down, the GLBP group weighting is reduced by a specified value.

By default, the GLBP virtual forwarder preemptive scheme is enabled with a delay of 30 seconds. A backup virtual forwarder can become the AVF if the current AVF weighting falls below the low weighting threshold for 30 seconds

Hope to help

If helpful do rate the post

Ganesh.H

johnnylingo · ‎02-05-2010

OK - got it. I agree that there is a 30 second delay between the tracking change and the AVF change:

Feb 5 20:04:52 UTC: %TRACKING-5-STATE: 1 interface Se0/0 line-protocol Up->Down
Feb 5 20:05:24 UTC: %GLBP-6-FWDSTATECHANGE: Fa0/0 Grp 10 Fwd 1 state Active -> Listen

Once the AVF shifts to the other router, traffic patterns start to change. However, it still takes about 3 minutes for traffic to entirely bleed off.

That's fine for my purposes - Just curious why it was taking longer than HSRP.

sachinraja · ‎02-05-2010

Hi Johny

What is the present hello/dead time intervals? Did you try tuning the timers ? you can probably try a 1 sec hello to see how fast your glbp converges.. the hold timer will then be 3 secs.. is there any other complications on your network ? Layer 3 convergence is also affected by STP convergence, routing protocol convergence etc, which you might need to take care of.. even if your GLBP converges, if your routing protocol doesnt converge, your backup routes might take x secs before IP routing table updates.. have a look at all your layer 2 and 3 protocols in use, and then decide..

Hope this helps.. all the best

Raj

Malik Nouman Ahmad · ‎04-25-2010

Hi Johny,

The command that I believe you are missing the following configuration in both routers

glbp 10 preempt

Although the decrement is happening but the other router is not taking over as AVG because preemption is not configured

I hope this solves your problem

Regards,

Malik

I realized that I was wrong as preemption for selection of AVG where in your case you want to keep the AVG as it is but want to shift the whole traffic to R2.

Actually your configuration should work