[j-nsp] SNMP traps for exceeding policer configuration

[Stefan Fouant]

On Wed, Feb 25, 2009 at 5:00 PM, Daniel Verlouw <daniel at bit.nl> wrote:
> On Feb 25, 2009, at 9:34 PM, Stefan Fouant wrote:
>>
>> I'd like to tragger some sort of alert when the traffic exceeds my
>> policer configuration and packets start being discarded.  I looked
>> through JUNIPER-FIREWALL-MIB and didn't see anything along the lines
>> of what I'm looking for.
>>
>> Anyone else implement anything along similar lines?
>
> have a look at
> enterprises.juniperMIB.jnxMibs.jnxFirewalls.jnxFirewallCounterTable.jnxFirewallCounterEntry
> (1.3.6.1.4.1.2636.3.5.2)
>
> E.g.:
>
> daniel at jun1.lab> show configuration interfaces xe-1/0/0.189 family inet
> policer
> input po-xe-1/0/0.189;
>
> daniel at jun1.lab> show policer | match po-xe-1/0/0.189
> po-xe-1/0/0.189-xe-1/0/0.189-log_int-i               170604
>
> using SNMP:
>
> jnxFWCounterDisplayFilterName.38.112.111.45.120.101.45.49.47.48.47.48.46.49.56.57.45.120.101.45.49.47.48.47.48.46.49.56.57.45.108.111.103.95.105.110.116.45.105.38.112.111.45.120.101.45.49.47.48.47.48.46.49.56.57.45.120.101.45.49.47.48.47.48.46.49.56.57.45.108.111.103.95.105.110.116.45.105.3 =
> po-xe-1/0/0.189-xe-1/0/0.189-log_int-i
>
> jnxFWCounterDisplayName.38.112.111.45.120.101.45.49.47.48.47.48.46.49.56.57.45.120.101.45.49.47.48.47.48.46.49.56.57.45.108.111.103.95.105.110.116.45.105.38.112.111.45.120.101.45.49.47.48.47.48.46.49.56.57.45.120.101.45.49.47.48.47.48.46.49.56.57.45.108.111.103.95.105.110.116.45.105.3 =
> po-xe-1/0/0.189-xe-1/0/0.189-log_int-i
>
> jnxFWCounterPacketCount.38.112.111.45.120.101.45.49.47.48.47.48.46.49.56.57.45.120.101.45.49.47.48.47.48.46.49.56.57.45.108.111.103.95.105.110.116.45.105.38.112.111.45.120.101.45.49.47.48.47.48.46.49.56.57.45.120.101.45.49.47.48.47.48.46.49.56.57.45.108.111.103.95.105.110.116.45.105.3 =
> 170604
>
> Works like a charm.
>
>   --Daniel.
>
>

Daniel - thanks, the jnxFWCounterPacketCount OID actually only has the
contents of the counter, what I want is actually the realtime bps
variable, so I guess I need to get that from the Interface MIB.

And I am assuming instead of trapping, we'll just need to regularly
monitor (poll) the given OID with our tool of preference and use that
to alert on.

-- 
Stefan Fouant

Windows XP crashed.
I am the Blue Screen of Death.
No one hears your screams.