[j-nsp] Netconf for VCP - EX4550

[Diogo Montagner]

The problem of this approach is that you are assuming all packets will have
the same size.

You could use the timestamp of each collection to calculate how many
packets traversed the interface in that period. It doesn't have a good
accuracy but will give something very close.

What about getting the pps and bps from the show interface queue vcp-x/y/z ?

Thanks

On Saturday, 17 August 2013, Giuliano Medalha wrote:

> People,
>
> We are doing a custom management system for VCP interfaces monitoring using
> NETCONF in JUNOS.
>
> The following stats are the only one reported to us:
>
> T2
> <input-packets>3966099</input-packets>
> <output-packets>3985901</output-packets>
>
> T1
> <input-packets>3966000</input-packets>
> <output-packets>3985850</output-packets>
>
>
> How is possible to calculate the input rate and the ouput rate ?
>
> The MTU for the interface is 1514.
>
> Can we consider something like:
>
> 1514 x 8 x (3966099-3966000) / (T2 - T1)
>
> It can be considering a goog mean value ?  Or we will have so much
> variations for packet size  ?
>
> packet size for JUNOS (VCP interface) is the same as ethernet frame-size ?
>
> Thanks a lot,
>
> Giuliano
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net <javascript:;>
> https://puck.nether.net/mailman/listinfo/juniper-nsp
>


-- 
./diogo -montagner
JNCIE-SP 0x41A