2 Replies Latest reply on Jan 23, 2013 10:37 AM by jamestrkcunningham

    Cisco 4000/4500 backplace utilisation? Using Cisco formulas?

    jamestrkcunningham

      I am trying to get a figure for the backplane utilisation of a Cisco 4000/4500 series switch.

      I know the STACK MIB is not available so querying the Cisco website I see a formula has to be used with 3 pieces of information from the mgmt MIB.

       

      Below shows the location of the 3 pieces of information.

      .1.3.6.1.2.1.2.2.1.10 ifInOctets

      .1.3.6.1.2.1.2.2.1.16 ifOutOctets

      .1.3.6.1.2.1.2.2.1.5 ifSpeed

       

      These are then included into a formula using the following guide;

      calculate_bandwidth_snmp.gif

      and then using this forumla;

      calculate_bandwidth_snmp3.gif

       

      In SolarWinds I have used UnDP to collect the 3 separate pieces of information and what I was wondering if anyone has managed to automate the formula (maybe in Report Writer or using Advanced SQL?) to gain a figure of throughput for these switches??? or another method that provides the same functionality?

       

      If so please please please share it

       

      James

        • Re: Cisco 4000/4500 backplace utilisation? Using Cisco formulas?
          mharvey

          This is close to the same formula that NPM uses to to covert the ifinOctets to bps then divide against ifSpeed to obtain the normal bandwidth utilitzation.  You would try this perhaps with a UnDP transform.  In theory if you have the ifinoctets and ifOutOctects setup as counter values, then NPM is only storing the delta anyway.  You could then do a transform poller of (({ifInOctets}*8)*100)/(600*{ifSpeed}) granted the 600 would be if you are polling stats every 10 minutes.  However, with these being RFC MIBs then Orion is already doing ifInOctests delta *8/stat poll time in number of seconds to get bps.  So I'd imagine it's already doing the calculation you found for interface utilization as this would be a standard calculation using the RFC standard MIBs.

           

          Regards,

          Matthew Harvey

          Loop1 Systems