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.

How to get data from a report into NOC

Hi everyone - I've got a question about getting relevant information into my NOC view for a service desk.

I've got a number of virtual servers that I'm monitoring. I can get volume information, and application monitor views, but when it comes to CPU utilisation, the resources I have available to me aren't very accurate and don't give a good indication of how much load the VM is experiencing. I really want load averages so I can see if a server is being overworked or is having an issue.

In the reporting section of Orion, I can run a report called "High VM CPU Utilization - Hour" which shows me the average CPU utilisation for the past hour on all my VMs where that average load is over 60%. How can I convert this value into something I can display in an NOC, and update it every 10 minutes?

I have a feeling that the figures reported here are far more accurate than the "Top 10" resources in the standard NOC views menu creation options.

pastedImage_0.png

Thanks for any assistance you can provide!

Cheers,

Gareth

  • Have you tried doing this in Pefstack? or do you have Virtualization Manager?

    pastedImage_0.png

    pastedImage_1.png

  • If the table from the report is something useful that would be beneficial to the NOC Team then I would suggest adding a Custom Table to your Dashboard and using the exact same formatting / Datasource that you have defined in your report. The report just contains a custom table.

  • Hi neomatrix

    Thanks for this, I need to show this information for around 55 VMs on one screen - is there a way to just display the VM name and the percentages, without the line graph? I've tried doing a custom table, but this doesn't really highlight when CPU usage is high, it just lists the CPU usage of specified VMs.

    Cheers,

    Gareth

  • Hi Gareth,

    A custom table or Custom Query would be your best result - you just need to tune your data source or use SQL/SWQL. What exactly do you want the table to look like?

    VM Name     |     Last Hour AVG     |     Memory Usage     |     CPU Load

    If you used Custom SQL/SWQL you could use a Case statement that changes a status indicator from Red/Yellow/Green depending on the threshold you have set or you could show the Warning/Critical icon if the CPU has reached a certain value or simply sort by the highest at the top!

  • Hi David,

    That sounds more like what I want to do - don't suppose you've got any links to resources on how to achieve this?

    Regards,

    Gareth

  • Just edit the page and select add widget.

    pastedImage_0.png

  • This is fairly rough, and if I had the time I would make it cleaner but it should at least get you started:

    SELECT

    vm.Name AS [Name]

    ,vm.DetailsUrl AS [_LinkFor_Name]

    ,CONCAT(ROUND(AVG(vm.VMStatistics.AvgCPULoad),2),' %') AS [AVG CPU Load]

    ,CONCAT(ROUND(vm.MemUsage,2), ' %') AS [Memory Usage]

    ,CONCAT(ROUND(vm.CpuLoad,2), ' %') AS [CPU Load]

    ,vm.Host.HostName AS [ESX Host]

    ,vm.Host.DetailsUrl AS [_LinkFor_ESX Host]

    ,vm.Host.Cluster.DisplayName AS [Cluster]

    ,vm.Host.DataCenter.DisplayName AS [DataCenter]

    ,vm.Host.DataCenter.DetailsUrl AS [_LinkFor_DataCenter]

    FROM Orion.VIM.VirtualMachines vm

    WHERE vm.GuestState = 'Running'

    GROUP BY vm.Name, vm.MemUsage, vm.CpuLoad, vm.Host.HostName, vm.Host.Cluster.DisplayName, vm.Host.DataCenter.DisplayName, vm.Host.DetailsUrl, vm.Host.DataCenter.DetailsUrl, vm.DetailsUrl

    Although I suspect that a Custom Table might look cleaner I will have to test it. Take out the Cluster / DataCenter as applicable.

  • Thanks David,

    This has been really helpful.

    Cheers