Conference netcad::hub_mgnt

Olli,

you should be able to see the errors by looking at HUBWatch repeater summary.
By selecting the port summary (click on the port), you will see the repeater
counters by clicking on the DETAILS button.
Use HUBwatch help for explanations on the counters.

You can also look at the hub900 extended repeater and rfc1516 mibs files to get
a description of the variables. You may use a mib browser to look at rfc1516
repeater mib  rptrMonitorPortInfo.rptprMonitorPortTable to get further details
on the exact errors. You have probably FCS, runts...


erptrTotalRptrErrors OBJECT-TYPE
    SYNTAX  Counter
    ACCESS  read-only
    STATUS  mandatory
    DESCRIPTION
            "The total number of errors which have occured on all
             the groups in a repeater. This object is a summation
             of  the values of the rptMonitorGroupTotalErrors as
             defined in RFC 1516 for all of the groups in a repeater."
    REFERENCE
            "Reference RFC 1516 repeater MIB"
    ::= { erptrRptrInfo 6 }

The mibs are available on the NPBU Web under Products and Technology , Technical
Data , HUB mibs. (URL is http://www.networks.europe.digital.com for Europe).
They are also available through anonymous ftp on gatekeeper under 
/pub/DEC/hub900/mibs

Hope This helps,

Manuel.



 

    
    
    Manuel,
    
    Thanks alot for your reply! I'll check those sources.
    
    
    Thanks,
    
    	-Olli
    

Olli,

I've been asking around about "erptrTotalRptrErrors".  A definition from
one of the developers is:

>The erptrTotalRptrErrors MIB object is a summation of all of the following
>errors which have been detected on all of the ports in a repeater:
>
>        FCS Errors
>        Alignment Errors
>        Frame Too Longs
>        Short Events
>        Late Events
>        Very Long Events
>        Data Rate Mismatches
>
>Note that this list does not include runts, auto-partitions, collisions, or
>health text changes.  This object is basically a summation of the
>rptrMonitorGroupTotalErrors object for all of the groups in the repeater,
>which in turn is a summation of the rptrMonitorPortTotalErrors for all the
>ports in a given group.  

I was surprised that there was an alarm on this object.  I claim that
default alarms  on very significant events like ports partitioning, etc., 
and not on normal network events like this.  The person who lobbied for 
this default claims:

>                     In  my mind the type of errors seen below can have a
>        more destructive effect on a  network  than a partitioned port (a
>        partitioned  port  is  partitioned  and  therefore   out  of  the
>        picture).  Things like late events can cause a network to be more
>        flaky and if you don't know what you  are  looking  for you won't
>        diagnose it that easily.
 
So I have a few questions for you.

   1)  Have you found the source of these errors, and if so, what was it?

   2)  Did you find the alarms useful, or a pain? 

Anyone else have an opinion on this alarms or alarms like it?

							-Shawn

    
    Shawn,
    
    	Thanks alot for your information. I really learn quite
    	alot about repeter related MIB's and about RMON events
    	and alarms when I read all those RFC's and used
        MIB browser.
    
    	FCS errors were incrementing total repeater errors.
    	especially AUI port. We haven't fix this error yet.	
    
    	It was good to see there is some error but after I found
    	how it works I disable total repeater error RMON alarm
    	and now we monitor port errors with SnmpCollector of
    	NETview. If error rate is over 0.5% in any repeater port
    	we generate event.
    
    	I am still little bit "confused" about  all this new stuff I have
    	learn. I have to think more about it before I give
    	more feedback. But any way it's very good to have RMON
    	alarm/event features. 
    	
    	I will post example how do you use NEtview (snmpset)
    	to modify RMON alarms.         
    
    	-Olli