[j-nsp] BGP Peer Route Table Size OID

[Pekka Savola]

On Mon, 12 Mar 2007, Scott Weeks wrote:
> snmpwalk -Os -c passwd -v 1 ipaddress 1.3.6.1.4.1.2636.5.1.1.2.6.2.1.7
>
> jnxExperiment.1.1.2.6.2.1.7.0.1.1 = Gauge32: 210210
> jnxExperiment.1.1.2.6.2.1.7.2.1.1 = Gauge32: 1
> jnxExperiment.1.1.2.6.2.1.7.3.1.1 = Gauge32: 1012
> jnxExperiment.1.1.2.6.2.1.7.4.1.1 = Gauge32: 209974
> jnxExperiment.1.1.2.6.2.1.7.5.1.1 = Gauge32: 11
> jnxExperiment.1.1.2.6.2.1.7.6.1.1 = Gauge32: 2
> jnxExperiment.1.1.2.6.2.1.7.7.1.1 = Gauge32: 0

I'll use this as a soapbox to complain about Juniper's implementation 
of jnxBgpM2PrefixInPrefixes{,Rejected,Accepted}.  Rejected shows the 
number of non-active routes from a peer.  Accepted shows the number of 
active routes from the peer.  There is no way to find out the number 
of routes which were rejected due to inbound policy.

The definition is:

     jnxBgpM2PrefixInPrefixesRejected OBJECT-TYPE
         SYNTAX     Gauge32
         MAX-ACCESS read-only
         STATUS     current
         DESCRIPTION
             "The number of prefixes for a peer that are installed
              in the Adj-Ribs-In and are NOT eligible to become active
              in the Loc-Rib."
         ::= { jnxBgpM2PrefixCountersEntry 9 }

.. which shows that the JunOS BGPv2 MIB implementation doesn't even 
conform to its own description (which is derived from the expired 
raft-ietf-idr-bgp4-mibv2-xx).  That is, 'NOT eligible to become active 
in the Loc-Rib' should not include non-primary (inactive) paths which 
may well become active if another route were to disappear. 
InPrefixesRejected is clearly meant to show the filtered routes, but 
alas, the current implementation does not.

-- 
Pekka Savola                 "You each name yourselves king, yet the
Netcore Oy                    kingdom bleeds."
Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings