cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
795
Views
0
Helpful
1
Replies

Alert when a BGP route changes to static?

Willie Walton
Level 1
Level 1

We have a dual MPLS network. I have multiple alerts configured through tracking objects and event manager applets. We receive routes from MPLS1 on router 1, and routes from MPLS2 go to router 2. There is also a link between the two routers so that if we lose a route on router 1, weighted static routes will send the traffic over to router 2 instead.

 

I have tracking objects configured to monitor reachability to each destination subnet on the MPLS, but the tracking object never goes down when the weighted static route is installed in the routing table.

 

Relevant configuration:

 

track 1 ip route 10.150.10.0 255.255.255.0 reachability

ip route 10.150.10.0 255.255.255.0 10.250.250.2 250

 

event manager applet PRIMARY-REACHABILITY-DOWN
  event track 1 state down

  (another config line that sends an email to our mail relay)


event manager applet PRIMARY-REACHABILITY-UP
  event track 1 state up
  (another config line that sends an email to our mail relay)

 

The "show track" information is listed below. When the BGP route goes away, the reachability state stays up, but it changes from "(BGP)" to "(static)"

 

R1#show track 1
Track 1
IP route 10.150.10.0 255.255.255.0 reachability
Reachability is Up (BGP)
25 changes, last change 20w6d
First-hop interface is GigabitEthernet0/0/1
Tracked by:
Static IP Routing 0
EEM 2495244088
EEM 2495245448

 

 

R1#show track 1
IP route 10.150.10.0 255.255.255.0 reachability
Reachability is Up (static)
25 changes, last change 20w6d
First-hop interface is GigabitEthernet0/0/2
Tracked by:
Static IP Routing 0
EEM 2495244088
EEM 2495245448

 

Is there any way to configure tracking to pay attention to the change from BGP to static and vice versa?

1 Accepted Solution

Accepted Solutions

Willie Walton
Level 1
Level 1

I have figured out a workaround for this by configuring an EEM applet to check for a change in transmit bandwidth utilization on our backup router's interface connected to the MPLS. Normally there is zero traffic on the secondary router's MPLS-connected interface outside of BGP, ARP, etc. So the transmit bits per second counter on the interface is constantly at 0. If router 1's route to the MPLS disappears, then it's going to send the traffic to router 2, and then router 2 will send the traffic to the MPLS. 


30 second input rate 0 bits/sec, 0 packets/sec
30 second output rate 0 bits/sec, 0 packets/sec

When I send a ping to the next layer 3 hop:

 

30 second input rate 3000 bits/sec, 4 packets/sec
30 second output rate 3000 bits/sec, 4 packets/sec

 

So I can monitor either transmit or receive bandwidth changes. Either will work. 

event manager applet MPLS-FAILOVER-BANDWIDTH
event interface name GigabitEthernet0/0/1 parameter transmit_rate_bps entry-op ge entry-val 1 entry-type value exit-op eq exit-val 0 exit-type value poll-interval 10
action 1.0 syslog priority notifications msg "MPLS HAS FAILED OVER TO THE BACKUP CIRCUIT"
action 2.0 mai ser "$_email_server" to "$_email_to" fro "$_email_from" sub "text" body "text"

View solution in original post

1 Reply 1

Willie Walton
Level 1
Level 1

I have figured out a workaround for this by configuring an EEM applet to check for a change in transmit bandwidth utilization on our backup router's interface connected to the MPLS. Normally there is zero traffic on the secondary router's MPLS-connected interface outside of BGP, ARP, etc. So the transmit bits per second counter on the interface is constantly at 0. If router 1's route to the MPLS disappears, then it's going to send the traffic to router 2, and then router 2 will send the traffic to the MPLS. 


30 second input rate 0 bits/sec, 0 packets/sec
30 second output rate 0 bits/sec, 0 packets/sec

When I send a ping to the next layer 3 hop:

 

30 second input rate 3000 bits/sec, 4 packets/sec
30 second output rate 3000 bits/sec, 4 packets/sec

 

So I can monitor either transmit or receive bandwidth changes. Either will work. 

event manager applet MPLS-FAILOVER-BANDWIDTH
event interface name GigabitEthernet0/0/1 parameter transmit_rate_bps entry-op ge entry-val 1 entry-type value exit-op eq exit-val 0 exit-type value poll-interval 10
action 1.0 syslog priority notifications msg "MPLS HAS FAILED OVER TO THE BACKUP CIRCUIT"
action 2.0 mai ser "$_email_server" to "$_email_to" fro "$_email_from" sub "text" body "text"