[c-nsp] AS53xx and SNMP OID modems-DS0s in use - bug

[Tassos Chatzithomaoglou]

Hi Robert,

This is a known bug (CSCej89068 if my notes are right)...

12.3(18.8) solves it, so i guess you must wait for 12.3(19) (and of May if i'm not wrong) or ask it 
from tac ;)
This is the reason we keep our AS5300s in 12.2(15)T16 till now...


--
Tassos

Robert Blayzor wrote on 18/4/2006 7:17 μμ:
> We have several AS53xx access servers deployed.  For well over a year
> we've been running IOS 12.3(x) without a single problem, the last
> version being 12.3(13a) which seems to be fine.
> 
> We've recently upgraded a few to 12.3(17a) and now 12.3(18) (mainline).
>  Since then we've noticed that the SNMP OID's used to gather the number
> of DS0's and/or modems does not work correctly.
> 
> The OID in question: 1.3.6.1.4.1.9.10.19.1.1.4.0
> 
> On our AS5300's still running 12.3(13a) this works as normal and the
> value returned in SNMP is accurate with the number of active calls on
> the router.  On one of the access servers running 12.3(13a) with a fair
> amount of uptime I get the proper response:
> 
> snmpget -c public x.x.x.x 1.3.6.1.4.1.9.10.19.1.1.4.0
> SNMPv2-SMI::enterprises.9.10.19.1.1.4.0 = Gauge32: 116
> 
> 
> Now on one with about a month of uptime running 12.3(17a):
> 
> snmpget -c public x.x.x.x 1.3.6.1.4.1.9.10.19.1.1.4.0
> SNMPv2-SMI::enterprises.9.10.19.1.1.4.0 = Gauge32: 5249
> 
> 
> So on the new version it seems that this number slowly creeps up,
> incrementing over time to where the number grows in the thousands.  It's
> almost as though the value misses a decrement internally when a call
> disconnects.  When this happens all of our MRTG collectors start storing
> numbers that are out of bounds.
> 
> I've done some searching but can't seem to find if it's a bug.  I'm not
> exactly sure which version between 12.3(13a) and 12.3(17a) actually
> broke it, but I do know that it's broken in 12.3(18) as well.
> 
> Anyone else run into this?
>