This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Call Manager Appliance (Cisco MCS 7845H) CPU Monitoring not working

I have seen the question before on the forums, but nobody has really given a difinitive answer. For some reason, my Call Manager appliances CPU and memory usage are not showing up in Solarwinds. I have found an OID that I have created a custom poller for but it seperates out each CPU individually. It would be nice to have the single averaged number in the gauge.

The OID I found is 1.3.6.1.2.1.25.3.3.1.2. Does anybody have any idea why Solarwinds doesn't collect the CPU and Memory data inherently? CPU and Memory do show up when I "list resources" but not on the node details page.

p.s. I can see the physical memory usage if I add the volumes wigit

  • Same problem here.  Did you ever find a solution?

  • yes same issue here.  I have a number of cucm boxes and none are showing cpu or mem?

    Dont really want to use UDPs either until Solarwinds allow us to control the polling frequency of a UDP. example - constantly polling every 120secs for a serial number is daft, multiply this by hundreds and you can see where this is going.

  • The issue is that the CCM

    -          Is a cisco device


    -          And supports the Host Resource MIB that has CPU and Memory, which NPM can work with.


     

    So why doesn’t it work?

     

    This is because it’s a Cisco device, therefore we expect CPU and memory to come from where they normally come,  for Cisco Routers and Switches.

    But it happens that a CCM is a server, so it does not report on CPU/memory like a router, but like a server, based on the Host Resource MIB.

     

    So basically we are ignoring the fact that it has Host Resource MIB (even if we suport this MIB), because for us it’s before anything else, a Cisco, so it should “behave like a router”, MIB-wise.

     

    So it’s not really a bug, but we definitely need to improve the way we categorize devices and go for the MIBs that are actually being supported.

     

    We are looking at this for a future release (65220)

     

    The UnDP as a temporary workaround, is the only alternative for now.

    This might help: