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.

Can we Monitor Cisco Server Hardware Sensor in Solarwinds ???

Hi Team,

we have cisco UCS A class and B class servers available. Can we monitor cisco severs hardware sensor in solarwinds ??

If yes --> How monitor hardware sensor in solarwinds. please guide to me.

Note : we don't need monitor traps. we have monitor exact cisco nodes hardware monitoring.

Please guide to me or provide any link that will helpful..

NPM module which version support for cisco servers hardware sensors. if you have could you provide MIB's

Thanks and Regrads,

Kishore

  • kishore574​, you should be able to monitor Cisco UCS hardware with NPM.

    Some paths to take, ideas to investigate for you:

    • Configure the UCS chasses to support Read-only snmp-v3 strings.
    • Set access control list rules in the chasses to only allow snmp-v3 access from your Solarwinds polling engines
    • Disable all snmp-v1 and snmp-v2c access to the UCS chasses
    • Add the UCS chasses as nodes to NPM using the snmp-v3 strings you built
    • Add the appropriate pollers to the newly-added UCS nodes.  There are a few different ways to do this--either add the pollers as you add UCS nodes, or add all UCS nodes and then apply the pollers simultaneously to those UCS nodes by using the Settings > Manage Nodes view, grouping on the left by Machine Type, selecting all UCS chasses, then using the "Assign Pollers" option on the right side.

    pastedImage_2.png

    From there select all pollers you wish to use for monitoring the UCS chasses.

    pastedImage_3.png

    If you do not find the UCS specific pollers you wish to use, you may wish to download and test any that other Thwack Users have shared.  To do so, open NPM's Settings > All Settings page.  Go to Node & Group Management > Manage Pollers.

    You'll see two tabs--Local Poller Library and Thwack Community Pollers.  Select either one and use the Search box at the upper right to look for the pollers that will fill your needs.

    pastedImage_5.png

    pastedImage_7.png

    pastedImage_4.png

    Apply the pollers to the appropriate nodes, making certain the pollers are enabled, and start testing to see if you get what you need.

    If the pollers already available do not suffice for your needs, follow the instructions for performing OID discoveries and MIB walks on your chasses.  Parse through the responses and then use the instructions for building a custom Universal Device Poller (UnDP) to achieve your goals.

    Some links to help you accomplish successful outcomes:

    It's easy to search for anything you might need in Thwack.com by entering the topic in the Search window:

    pastedImage_13.png

    If you still don't find what you need, open a Support Case with SolarWinds and explain what you need and how it's different from what you've found available.  They'll help you get going in the right direction.

    And if you find you're still interested in reading the UCS MIBs, you can find them from Cisco via Google:

    https://www.google.com/search?q=cisco+UCS+MIB&rlz=1C1GCEU_enUS819US819&oq=cisco+UCS+MIB&aqs=chrome..69i57j0l5.7991j0j4&s…

    Swift Packets!

    Rick Schroeder

  • Hi rschroeder ,

    Thanks for update. we are thinking to monitor these Cisco UCS servers with WMI polling method instead of SNMP, is it possible to monitor hardware sensors.

  • My team does not monitor the UCS chasses themselves--that is the responsibility of another team--the Systems Analysts.  They do not use snmp to monitor UCS chassis hardware.  Their solution is NRPE and is completely agent-based.  They tie it into Nagios since that's their area of comfort.

    My team uses WMI to monitor multiple Windows 2016 servers that are VMs hosted within those UCS chasses.  It works well for the servers.

    Using WMI to manage servers hosted inside those UCS chasses allows us to gather a variety of information about the servers without using snmp, including:

    • Alerts
    • Latency
    • IP addresses
    • Availability
    • CPU utilization and Capacity Forecasts
    • Packet loss
    • Events (Summary)
    • Node details that include:
      • Machine Type
      • System descriptions (Intel family and models, software versions, etc.)
    • Min/Max/Average Memory usage over any time period
    • Netflow conversations to/from the host
    • Disk Volumes
    • Percent disk space used
    • Interfaces & errors on them
  • You may want to check out the SAM RC2 post - we've actually done some work here on enhanced UCS support in SAM SAM 6.8 RELEASE CANDIDATE 2 NOW AVAILABLE!