GLBP configuration

[sly007]

Kindly assist in troubleshooting GLBP functionality

GLBP has been configured one 2951 and 3945 (using the etherswitch module on 3945) Routers, using OSPF.   After configuration, the load-balancing is working properly, as the route from the local systems periodically changes from Route 1 to the other and back at regular interver. But whenever I shut down a link, to test the functionality and availability of the redundant link of GLBP, the whole  circuit would go down, making destination no longer reachable. I believe there is something that ought to be done or some of the configurations is not complet. Please help troubleshoot, the configs are below.

In addition, how is the "ip sla group"  configured on the 2951 and 3945 Routers, to be able to track objects? It gives me the "ip sla responder" and "ip sla key-chain" options on image version c2951-universalk9-mz.SPA.150-1.M3.bin

ROUTER_1

GigabitEthernet0/0         192.168.215.252 YES NVRAM  up                    up

GigabitEthernet0/1         192.168.155.2   YES NVRAM  up                    up

GigabitEthernet0/2         unassigned      YES NVRAM  administratively down down

GigabitEthernet1/0         unassigned      YES NVRAM  administratively down down

Router_1#sh run int g0/0

interface GigabitEthernet0/0

description Call Center LAN

ip address 192.168.215.252 255.255.254.0

duplex auto

speed auto

glbp 10 ip 192.168.215.254

glbp 10 timers 5 18

glbp 10 priority 101

glbp 10 preempt

glbp 10 weighting 100 lower 91 upper 91

glbp 10 load-balancing weighted

glbp 10 weighting track 1 decrement 10

track 1 interface GigabitEthernet0/1 line-protocol

ROUTER_2

Interface                  IP-Address      OK? Method Status                Protocol

GigabitEthernet0/0         unassigned      YES NVRAM  administratively down down

GigabitEthernet0/1         192.168.155.6   YES NVRAM  up                    up

GigabitEthernet0/2         unassigned      YES NVRAM  down                  down

Serial0/0/0:0              unassigned      YES NVRAM  down                  down

Serial0/0/1:1              unassigned      YES NVRAM  down                  down

GigabitEthernet2/0         192.168.215.253 YES NVRAM  up                    up

Router_2#sh run int g2/0

interface GigabitEthernet2/0

ip address 192.168.215.253 255.255.254.0

ip nbar protocol-discovery

glbp 10 ip 192.168.215.254

glbp 10 timers 5 18

glbp 10 preempt

glbp 10 weighting 100 lower 91 upper 91

glbp 10 load-balancing weighted

glbp 10 weighting track 1 decrement 10

track 1 interface GigabitEthernet0/1 line-protocol

[John Blakley]

It looks like you're tracking the incorrect interface. You should be tracking g0/0 in order for the decrement to work. On R2, you're tracking the incorrect interface there as well, so you'd want to change it to g2/0.

As far as your other question, the Universal base image doesn't support a full implementation of ip sla. You'd need at least the data license for that functionality. But overall, your tracking command should work.

HTH,

John

[sly007]

Hello John,

Please correct me if i am wrong. I was thinking the uplink/wan interface is the one to track..

The G0/0 interface is the LAN interface, and it is not expected to break down, because that is where the GLBP is configured.

However, could that be the only reason why the local destination would not be reachable when one of the links goes down?

Please reply.

Thank you.

[John Blakley]

My apologies. So, since you're tracking the wan interface, did you physically pull the cable to simulate an outage? One problem with tracking on an ethernet segment is that if your circuit were to go down, you won't necessarily see your line protocol go down because it's connected into a switch that's still up. This is where the sla comes in. You can ping the other side and then fail over if the other side stops responding.

So, a couple of questions:

1. How are you testing?

2. Can you test again?

3. When it's down, can you post "show glbp"?

John

[sly007]

Thanks for your prompt response.

1. I test by administratively shutting down the end of the WAN interface
2. I just tested before i sent the reply, and
3. below is the output when i just shut down the uplink interface again.

ROUTER_1

GigabitEthernet2/0 - Group 10

  State is Standby

    1 state change, last state change 01:27:50

  Virtual IP address is 192.168.215.254

  Hello time 5 sec, hold time 18 sec

    Next hello sent in 0.928 secs

  Redirect time 600 sec, forwarder timeout 14400 sec

  Preemption enabled, min delay 0 sec

  Active is 192.168.215.252, priority 101 (expires in 17.856 sec)

  Standby is local

  Priority 100 (default)

  Weighting 100 (configured 100), thresholds: lower 91, upper 91

    Track object 1 state Up decrement 10

  Load balancing: weighted

  Group members:

    c84c.7563.8750 (192.168.215.253) local

    f866.f27b.4db0 (192.168.215.252)

  There are 3 forwarders (1 active)

  Forwarder 1

    State is Listen

    MAC address is 0007.b400.0a01 (learnt)

    Owner ID is unknown

    Time to live: 8959.360 sec (maximum 8962 sec)

    Preemption enabled, min delay 30 sec

    Active is 192.168.215.252 (secondary), weighting 100 (expires in 15.936 sec)

  Forwarder 2

    State is Listen

    MAC address is 0007.b400.0a02 (learnt)

    Owner ID is f866.f27b.4db0

    Time to live: 14397.376 sec (maximum 14400 sec)

    Preemption enabled, min delay 30 sec

    Active is 192.168.215.252 (primary), weighting 100 (expires in 18.304 sec)

  Forwarder 3

    State is Active

      1 state change, last state change 01:27:55

    MAC address is 0007.b400.0a03 (default)

    Owner ID is c84c.7563.8750

    Preemption enabled, min delay 30 sec

    Active is local, weighting 100

ROUTER_2

GigabitEthernet0/0 - Group 10

  State is Active

    1 state change, last state change 01:34:45

  Virtual IP address is 192.168.215.254

  Hello time 5 sec, hold time 18 sec

    Next hello sent in 0.096 secs

  Redirect time 600 sec, forwarder timeout 14400 sec

  Preemption enabled, min delay 0 sec

  Active is local

  Standby is 192.168.215.253, priority 100 (expires in 17.056 sec)

  Priority 101 (configured)

  Weighting 100 (configured 100), thresholds: lower 91, upper 91

    Track object 1 state Up decrement 10

  Load balancing: weighted

  Group members:

    c84c.7563.8750 (192.168.215.253)

    f866.f27b.4db0 (192.168.215.252) local

  There are 3 forwarders (2 active)

  Forwarder 1

    State is Active

      1 state change, last state change 01:30:35

    MAC address is 0007.b400.0a01 (learnt)

    Owner ID is c84c.7563.8718

    Redirection disabled

    Time to live: 8967.168 sec (maximum 14400 sec)

    Preemption enabled, min delay 30 sec

    Active is local, weighting 100

    Client selection count: 68

  Forwarder 2

    State is Active

      1 state change, last state change 01:34:09

    MAC address is 0007.b400.0a02 (default)

    Owner ID is f866.f27b.4db0

    Redirection enabled

    Preemption enabled, min delay 30 sec

    Active is local, weighting 100

    Client selection count: 500

  Forwarder 3

    State is Listen

    MAC address is 0007.b400.0a03 (learnt)

    Owner ID is c84c.7563.8750

    Redirection enabled, 599.072 sec remaining (maximum 600 sec)

    Time to live: 14399.072 sec (maximum 14400 sec)

    Preemption enabled, min delay 30 sec

    Active is 192.168.215.253 (primary), weighting 100 (expires in 17.504 sec)

    Client selection count: 460

Thank you

[John Blakley]

Can you also post "show track" while this is down?

[sly007]

please find below the result of "show track" while the line is shut down

Track 1

  Interface GigabitEthernet0/1 line-protocol

  Line protocol is Up

    1 change, last change 00:07:13

  Tracked by:

    GLBP GigabitEthernet2/0 10.

What could be wrong?

[skarthic]

Really interesting. So this o/p of 'show track' is when the interface gi0/1 was admin shut'd. You are sure the interface went down. Did you capture a 'show interface gi0/1' / 'show ip int br' just to confirm?

[sly007]

Thanks for your assistance.

The GLBP works quite well, providing Hardware Availability, and Load Balancing, at this I have confirmed this.

But the challenge is when any of the interfaces at the Headquarters is shut down, the whole link goes off. I expected to see a redundancy.

Please see the attached diagram in PDF.

I am thinking the reason while the tracking is "is not effective" is because the interface at the remote site is connected to an active port on the transparent switch. Therefore, I seek to configure "IP SLA object" on the 2951 and 3945 Routers.

Is there any solution on how to "IP SLA object" on Cisco 2951 and Cisco 3945 Routers?

[John Blakley · Accepted solution]

Do you get a different result if you physically pull the cable? If you have a specific route that you receive from the other side, you could try changing the tracking config to track the route. For example, if you receive prefix 192.168.1.0/24 from the provider, you could configure track like:

track 1 ip route 192.168.1.0/24 reachability

Otherwise, there are some tracking issues with the IOS version that you're running. You could try to update the IOS to see if this other problem is resolved. Personally, if pulling the cable still doesn't show you that the interface is down (tracked), then I would update the IOS first to see if it worked. I've tested shutting down the interface to see if line protocol was seen as down, and it is:

Track 1

  Interface FastEthernet0/0 line-protocol

  Line protocol is Down (hw admin-down)

    4 changes, last change 00:00:04

Notice that it says admin-down, so you, in theory, should be seeing the same thing. Your glbp configuration should work fine once you get the tracking situation resolved.

HTH,

John