vcenter Cluster report

[DevD]

I am looking for a script to run against the vcenter to generate the following table  send it in email body  (if not possible in body than as attached csv)

                                                                                                                                                                                

Hostname	No.of total VM's	Power-on VM's	Power-off VM	CPU Utilization	Memory Utilization
ABC
XYZ
ABC
XYZ

[LucD]

Try something like this

div>$report = Get-Cluster -PipelineVariable cluster |

Get-VMHost -PipelineVariable esx |

ForEach-Object -Process {

   $vms = $esx | Get-VM

   New-Object PSObject -Property ([ordered]@{

   Cluster = $cluster.Name

   VMHost = $esx.Name

   VMPoweredOn = ($vms | where{$_.PowerState -eq 'PoweredOn'}).Count

   VMPoweredOffn = ($vms | where{$_.PowerState -eq 'PoweredOn'}).Count

   CpuUtilPerc = (Get-stat -Entity $esx -Stat cpu.usage.average -Realtime -MaxSamples 1 -Instance '').Value

   MemUtilPerc = (Get-stat -Entity $esx -Stat mem.usage.average -Realtime -MaxSamples 1).Value

   })

}

$sMail = @{

   From = 'me@domain'

   To = 'you@domsain'

   Subject = 'Cluster report'

   Body = $report | ConvertTo-Html

   SmtpServer = smtp@domain

   BodyAsHtml = $true

}

Send-MailMessage @sMail

[DevD]

Hi Luc, Thanks for responding.

Script failed with following error

Get-stat : 4/8/2019 9:11:06 AM    Get-Stat        The metric counter "cpu.usage.average" doesn't exist for entity "XXXX.XXX.com".   

At D:\Scripts\Clusterreport.ps1:19 char:19

+ ... UtilPerc = (Get-stat -Entity $esx -Stat cpu.usage.average -Realtime - ...

+                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    + CategoryInfo          : ResourceUnavailable: (cpu.usage.average:String) [Get-Stat], VimException

    + FullyQualifiedErrorId : Client20_RuntimeDataServiceImpl_CheckUserMetrics_MetricDoesntExist,VMware.VimAutomation.ViCore.Cmdlets.Commands.GetViStats

Get-stat : 4/8/2019 9:11:06 AM    Get-Stat        Stat with instance '' was not found using the specified filter(s).   

At D:\Anoop\Scripts\Clusterreport.ps1:19 char:19

+ ... UtilPerc = (Get-stat -Entity $esx -Stat cpu.usage.average -Realtime - ...

+                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    + CategoryInfo          : ObjectNotFound: (:) [Get-Stat], VimException

    + FullyQualifiedErrorId : Core_OutputHelper_WriteNotFoundError,VMware.VimAutomation.ViCore.Cmdlets.Commands.GetViStats

Get-stat : 4/8/2019 9:11:06 AM    Get-Stat        The metric counter "mem.usage.average" doesn't exist for entity "XXXX.XXXX.com".   

[LucD]

Looks like you are not collecting statistical data on that ESXi node.
Which is rather bizar.

Do you see performance data for that ESXi node under the Performance tab?

[DevD]

Yeah you were right now no more than error message but seeing the below one :

Send-MailMessage : Cannot convert 'System.Object[]' to the type 'System.String' required by parameter 'Body'. Specified method is not supported.

At D:\Scripts\Clusterreport.ps1:49 char:18

+ Send-MailMessage @sMail

+                  ~~~~~~

    + CategoryInfo          : InvalidArgument: (:) [Send-MailMessage], ParameterBindingException

    + FullyQualifiedErrorId : CannotConvertArgument,Microsoft.PowerShell.Commands.SendMailMessage

[LucD]

Try replacnig that line with

Body = $report | ConvertTo-Html | Out-String

[DevD]

It worked ...Thanks!

is it possible to get the total at the end - especially for no of poweron/power-off VM's.

Cluster VMHost VMPoweredOn VMPoweredOffn CpuUtilPerc MemUtilPerc

HyperFlex Tes1 4 4 1.01 17.72

HyperFlex Test2 2 2 0.63 16.93

HyperFlex Test3 1 1 3.11 15.36

[LucD]

Try like this

$report = Get-Cluster -PipelineVariable cluster |

Get-VMHost -PipelineVariable esx |

ForEach-Object -Process {

   $vms = $esx | Get-VM

   New-Object PSObject -Property ([ordered]@{

   Cluster = $cluster.Name

   VMHost = $esx.Name

   VMPoweredOn = ($vms | where{$_.PowerState -eq 'PoweredOn'}).Count

   VMPoweredOff = ($vms | where{$_.PowerState -ne 'PoweredOn'}).Count

   CpuUtilPerc = (Get-stat -Entity $esx -Stat cpu.usage.average -Realtime -MaxSamples 1 -Instance '').Value

   MemUtilPerc = (Get-stat -Entity $esx -Stat mem.usage.average -Realtime -MaxSamples 1).Value

   })

}

$report += New-Object PSObject -Property ([ordered]@{

   Cluster = ''

   VMHost = ''

   VMPoweredOn = ($report | Measure-Object -Property VMPoweredOn -Sum).Sum

   VMPoweredOff = ($report | Measure-Object -Property VMPoweredOff -Sum).Sum

   CpuUtilPerc = ($report | Measure-Object -Property CpuUtilPerc -Average).Average

   MemUtilPerc = ($report | Measure-Object -Property MemUtilPerc -Average).Average

   })

$sMail = @{

   From = 'me@domain'

   To = 'you@domsain'

   Subject = 'Cluster report'

   Body = $report | ConvertTo-Html | Out-String

   SmtpServer = smtp@domain

   BodyAsHtml = $true

}

Send-MailMessage @sMail

[DevD]

It worked, need to find limit the average output to two dight after "." in sum.

Cluster VMHost VMPoweredOn VMPoweredOff CpuUtilPerc MemUtilPerc

HyperFlex Test1 4 4 1.23 17.72

HyperFlex Test2 2 1.03 16.93

HyperFlex Test3 1 3.38 15.36

                           7 7 1.88000003496806 16.6699997584025

[LucD · Accepted solution]

Change those 2 lines to

   CpuUtilPerc = [math]::Round(($report | Measure-Object -Property CpuUtilPerc -Average).Average,2)

   MemUtilPerc = [math]::Round(($report | Measure-Object -Property MemUtilPerc -Average).Average,2)

[DevD]

Perfect, thank you !!!

[DevD]

Hi Luc,

I notice the poweroff/poweron numbers are not valid - it seems it's duplicating the values

VMPoweredOn	VMPoweredOff	CpuUtilPerc	MemUtilPerc
17	17	0.01 0.01
18	18	21.46 25.81
18	18	25.15 24.81
19	19	21.71 25.33
18	18	22.91 26.92
18	18	25.78 24.81
18	18	26.24 24.83
18	18	20.41 25.84
18	18	26.74 25.85
18	18	29.5 24.82
18	18	16.95 25.89
18	18	15.46 24.84
18	18	13.94 25.86
234	234	20.48 23.51

[sjesse]

Find this part

VMPoweredOn = ($vms | where{$_.PowerState -eq 'PoweredOn'}).Count

   VMPoweredOff = ($vms | where{$_.PowerState -eq 'PoweredOn'}).Count

And change the second one to PoweredOff

[LucD]

There was indeed an error in the count for the powered off VMs (I put -eq instead of -ne).
I corrected the code above.

Changing it to PoweredOff will not count the Suspended VMs, that is why I used the -ne