[c-nsp] A few very Quick IP SLA questions

[Shimol Shah]

Inline

On 7/16/10 11:14 AM, Drew Weaver wrote:
> Hi all, happy Friday.
>
> A few questions regarding configuring IP SLA.
>
> I've configured two IP SLA probes as such:
>
> ip sla 1
>   icmp-echo x.x.25.97 source-ip x.x.25.98
>   frequency 10
> ip sla schedule 1 life forever start-time now
>
> ip sla 2
>   icmp-echo x.x.25.101 source-ip x.x.25.102
>   frequency 10
> ip sla schedule 2 life forever start-time now
>
> 1) If I want this probe to run forever, is it best to configure it as a recurring probe or have the lifetime be 'forever'?

I would do "forever"

>2) If the router has multiple paths to the destination does specifying 
the source-address mean that 100% of the time it will use the Interface 
that the indicated source address is assigned to?

No .. should be load-balanced.

Not related but beware of

CSCtf11508    IP SLA should not send probes when source-interface has no 
ip address



> 3) When using the 'track command' (for example: track 100 ip sla 1 reachability | state) What is the functional difference between reachability and state? Wouldn't they be the same thing?
>

Take a look at

http://www.cisco.com/en/US/docs/ios/ipapp/command/reference/iap_t1.html#wp1148364


> Also the main reason for implementing this is because we had an instance where a interface didn't go down, but no traffic would pass through it (routing protocols failed, etc) and we have our default routes setup as such:
>
> ip route 0.0.0.0 0.0.0.0 Vlan4091 x.x.25.97
> ip route 0.0.0.0 0.0.0.0 Vlan4092 x.x.25.101
>
> So return traffic was still being sent down the 'dead but up/up' interface which caused obvious heartache.
>
> Would using a track on each of these routes (combined with aforementioned IP SLA probes) be a good way to prevent this from occurring in the future?
>

Yes.


Shimol



> I basically want to ensure that both the interface is up and that traffic can pass from this router to its gateway before the route will be used.
>
> Sorry this is so long, hopefully it makes at least some sense.
>
> I thought about using BFD, but it seems like they have removed support for BFD on VLANs in recent code.
>
> Thanks,
> -Drew
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/