Steve Francis

LogicMonitor Staff
  • Posts

  • Joined

  • Last visited

Community Answers

  1. Steve Francis's post in How to play custom vedio in dashbaord was marked as the answer   
    If you have an iframe that the video clip can be referenced in, you can put it in an HTML widget on a dashboard.
  2. Steve Francis's post in host option to automatically add set of custom properties? was marked as the answer   
    This has been available for while - we apparently neglected to update this thread, though, sorry!
  3. Steve Francis's post in File Integrity Monitoring was marked as the answer   
    Nope. Not something we cover. We try to be excellent at performance, availability and capacity monitoring, alerting and presentation - which necessarily means we cant cover everything.
    That's one of the things we don't cover for now, sorry. Tripwire was what sprang to mind for me, too....
  4. Steve Francis's post in CPU metrics for - Dell EqualLogic FS7610 NAS appliance was marked as the answer   
    For the EqualLogic SANs - Dell does not expose the CPU or memory metrics at all. Dell's response is "No, there's no way to monitor the internal stats like CPU, cache, etc..  Given the design, a simple CPU/MEM graph wouldn't provide you with relevant information.   This is especially true in multimember groups.
    What you care about is how the array(s) are handling the IO load under the various circumstances. "
    Regarding the 7610 - if we aren't monitoring it adequately, contact your account manager to open a datasource request. (And if you get the MIBs from that would help.)
  5. Steve Francis's post in Why does one of my ESXi Servers show an uptime of -1 second? was marked as the answer   
    Generally, this means your vcenter is overloaded, or under configured. "-1" is vcenter shorthand for "I can't answer this right now". It can also be returned when querying a VM, when the VM has been powered off. You'll get -1 for a few samples, then NaN (Not a number).
    Anyway to correct: as noted
    You must ensure your Vcenter server is sized appropriately for the numbers of hosts and VMs it is managing. See VMware's documentation for size requirements.  (Note that the Vcenter database requirements must be added to the Vcenter server requirements if they are on the same machine.)  If you are using Vcenter 5.5 or lower, it is also often necessary to tune the Vcenter memory configuration, as documented in VMware's knowledgebase here and here. (VCenter 6 will dynamically adjust the memory allocation to the services if the VM running Vcenter is allocated more memory and rebooted.) If Vcenter does not have enough resources, it may refuse some connections to the API (HTTPS) port (easily seen in the HTTPS- datasource in LogicMonitor), or it may report  a value of "-1" to some performance queries.  Both of these situations will cause gaps in graphs. 
  6. Steve Francis's post in How to monitor Cisco UCS server hardware resources was marked as the answer   
    You can import the UCS datasources, which give you hardware monitoring among other things,  from the core repository.
    From Settings...Datasources...Add From LogicMonitor Repository.
    Sort by name, and check all the Cisco UCS ones to import.
  7. Steve Francis's post in Does LogicMonitor support Cisco IOS NetFlow? was marked as the answer   
    Yep. See
  8. Steve Francis's post in Is it possible to "acknowledge then ignore" an alert? was marked as the answer   
    Only interfaces with a status of active are discovered for monitoring - but they are not removed when they go inactive, so they keep alerting.
    In this case, as the disconnected port is not active, just delete the interface instance in logicmonitor. That will remove the alert.
    If the port ever gets connected to something again, it will be re-discovered, and monitoring will resume.
  9. Steve Francis's post in No Consolidation in Graphing was marked as the answer   
    The consolidation function only comes into play when looking at longer time periods.n
    nIf looking at a graph of 5 hours or less, then even if the data is collected every minute, no consolidation is used - all the individual datapoints are plotted. n
    nSo set your time scale for the graph in question to 1, 2 or 5 hours, and you get raw data.