EEM IP SLA Question

[Patrick McHenry]

Hi,           

I've configured an IP SLA and Event Manager to bring up a backup interface when the primary goes down. I have a ping to 4.2.2.2 for the IP SLA and when I shut the primary interface which, has a DMVPN to the Main site, the Event Manager is triggered and the secondary interface brings up the DMVPN with the Main site and all is good. When I try a different test where I cannot ping 4.2.2.2 but, the primary interface stays up(which is what will probably happen in the realworld) the Event manager is again triggered but the secondary interface never creates the tunnel back to the Main site. Both interfaces are configured with ip address dhcp and the default route is ip route 0.0.0.0 0.0.0.0 dhcp so, I'm wondering if this is my problem. When the primary is up, the secondary is down but, when the secondary is up I need the primary to also be up to continuosly ping 4.2.2.2.

I need this primary interface to stay up so it can continuosly ping 4.2.2.2 and  trigger the Event manager again when it is successful.

Is there a work around?

Thank  you, Pat

[cadet alain]

Hi,

just do a local PBR to always use the primary to do the IP SLA test.

Regards.

Alain

Don't forget to rate helpful posts.

[Patrick McHenry]

Alain,

I don't  think that is my problem.

My config for the IP SLA is:

track 1 ip sla 10 reachability
delay down 10 up 10
!
exit
ip sla 10
icmp-echo 8.8.8.8 source-interface GigabitEthernet 0/0
frequency 30
!
exit
ip sla schedule 10 life forever start-time now
!

So, it should always use that interface, no?

Or, could I create a static route to 4.2.2.2 from int g0/0. That way, the router will force that SLA ping out that interface?

Thanks, Pat.

[Patrick McHenry]

I just added the static route but, didn't help.

[Patrick McHenry]

I actually have it working at the moment but there is wierd going on. When I fail over to the backup interface(another DSL provider) by unplugging the voice cable to the primary DSL modem(that way, the g0/0-primary interface stays up and retains it's address),  the pings are not stable. Is the router getting confused because interface g0/0 is still up. I can send more config.

Protocol [ip]:

Target IP address: 173.194.43.39

Repeat count [5]: 1000

Datagram size [100]:

Timeout in seconds [2]:

Extended commands [n]:

Sweep range of sizes [n]:

Type escape sequence to abort.

Sending 1000, 100-byte ICMP Echos to 173.194.43.39, timeout is 2 seconds:

!!!!!!!!!!!!!!!!!..!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!.!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!.!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!

!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!.!!!!

!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!.!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!

!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!.!!!!!!!.!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!.!!!!!!.!!!!!!!!!!!!!!!!!.!!..!!!!!!!!!!!!!.!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!.!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!

!!!!.!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!.!!!!!!!

Thanks

[cadet alain]

Hi,

the only way to get the pings to always use the primary interface is to do a local PBR like I suggested, the source-interface

has nothing to do with the way traffic is switched.

Did you try it ? I should solve your problem normally.

Regards.

Alain

Don't forget to rate helpful posts.