[mrtg] Problem to get OID/MIB for collision packets

[MYCheng]

Hi Daniel J McDonald,

Good day!

Thanks for the advice to fix our IPSO firewall packet collisions issue,
however this only workable for firewall but in BIG-IP load balancer, it
didn't work, my efforts as follows:

> use the same MIB file ETHERLIKE-MIB to retrieve an information from
> BIG-IP-LB but failed.
>use the MIB file from MIBdepot site that grouped under F5 big-IP vendor
area, the correct MIB file to get this monitoring item
dot3StatsSingleCollisionFrames,dot3StatsMultipleCollisionFrames,dot3StatsExcessiveCollisions,dot3StatsLateCollisions)
is F5-BIGIP-SYSTEM-MIB, after put this MIB file into the GEtif folder, i can
see the symbolic name appeared in the GEtif list but while i trying to
retrieve the result remotely it was failed, no result shown, the snmp
connectivity is fine, just wondering why i can't retrieve the information.

Seeking your kind advise to fix this issue.

Thanks in advance.
 
Regards
Cheng - System Engineer.


Daniel J McDonald-2 wrote:
> 
> On Mon, 2008-11-17 at 01:31 +1200, alan cheng wrote:
>> Hello,
>> Currently i am in progress to setting up our MRTG to monitor our
>> Big-IP Load balancer and Nokia IPSO IP380 firewall, my problem is
>> while i looking for the Packet collision OIDs for these devices using
>> Getif MIB browser, i can't find any symbolic name or OID in these
>> devices, even though i try to do searching using a keyword such as
>> "packet" or "collision" in the MIBs text file, i still not able to
>> locate the OIDs which associated to this monitoring item,sound like
>> there is no packet collision OID existed in these devices, anyone have
>> any idea on how can i get a correct OIDs to monitor this item ? 
> 
> Why are you running half-duplex connections that would require
> collisions?  Full-duplex ethernet has worked pretty well since it was
> standardized in 1998.  And even if you were running half-duplex
> connections, why would you care about a flow-control method?  The myth
> that collisions were bad has been thoroughly debunked now that there are
> no token-ring sales-weasels[1] to stir the pot every now and again.
> 
> But, if the device is going to support the counters, it would be in the
> ETHERLIKE-MIB, with symbolic names of:
> dot3StatsSingleCollisionFrames,
> dot3StatsMultipleCollisionFrames,
> dot3StatsExcessiveCollisions,
> dot3StatsLateCollisions,
> 
> -- 
> Daniel J McDonald, CCIE #2495, CISSP #78281, CNX
> Austin Energy
> http://www.austinenergy.com
> 
> [1] - This is not to disparage weasels, which appear to be reasonably
> friendly rodents.
> 
>  
> _______________________________________________
> mrtg mailing list
> mrtg at lists.oetiker.ch
> https://lists.oetiker.ch/cgi-bin/listinfo/mrtg
> 
> 

-- 
View this message in context: http://n2.nabble.com/Problem-to-get-OID-MIB-for-collision-packets-tp1505815p1574804.html
Sent from the MRTG Users Mailinglist mailing list archive at Nabble.com.