Extremely slow when run PowerShell script by scheduled taks

I like automition/programming as much as I like Windows, my first offical training in University was VB, then I learned PHP and HTML after graduated. I can’t believe my first PHP program was welcomed in internel. But I stopped develop the skill.

I got lot of special request for virtual environment. Some people want to monitor snapshot, some want to check space status, and some request to create backup for VM…etc. Regular operation cannot satisify the requests, most of the requests are time consuming operation. Automation can save my ass. 🙂

You may know PowerCLI – VMware automation product for PowerShell. If you put PowerCLI scripts as scheduled task, you will be able to do any workflow you want by schedule.

But there is a problem stuggled me for a long time. The scheduled PowerCLI script is instable, sometimes script ran very well, sometimes script ran extremely slow, sometimes script hung, or sometimes script unexpected stopped. In troubelshooting I noticed it need a long waiting time (maybe 5 – 10 mintues) till execute first line of script when I ran it on Scheduled Task Window. Initially I thought it’s a bug of PowerShell 2.0. I upgraded to 3.0 but no lucky. I also suspected loading PowerCLI Snap-in may slow down the script exectue time, or bugs on PowerCLI components, tried to google but nothing was found.

Finally my colleague Amnon gave my some idea, he asked me change the scheduled taks priority. I figured out the trick after did some testing:

Create a task as usual.
Export the taks to a xml file.
Open the xml file by editor.
Search keyword priority.
Change the default value 7 to 3. ( More detail about priority of Scheduled Task )
Save the xml file and delete created task.
Import the xml as a task.

It signaficantly improved the executive time! Same script was 2 hours, now only 10 minutes!

Here is my another post ($array.count does not show anything in PowerCLI) regarding a bug of PowerShell, it may helpful for your PowerCLI script.

$array.count does not show anything in PowerCLI

I developed a script to take snapshot on VM and delete older snapshot regularly. I used variable $snapshots to receive data from Get-Snapshot command, then determine how many existing snapshots according to return of $snapshots.count. Somehow the script always ran incorrect behavior. When I deep looked into $snapshots, I found a interesting things.

Computer A, $snapshots.count return nothing if no, or one snapshot received.

Computer B, $snapshots.count returned 0 if no, or returned 1 if one snapshots received.

Then I checked out $PSVersionTable, found computer A powershell version is 2.0, B is 3.0, looks like this is a bug in 2.0.

The issue got fixed after upgrade to 3.0, one more benefit is I got performance improve to load PowerCLI components after upgrading.

You could install Widnows Management Framework 3.0 to upgrade PowerShell. I tested it on PowerCLI 5.5 Update1.

Device or Resource Busy

You may read my post How to find which ESXi 5.1 host lock the VM, it’s a solution to figure out which host lock down a file.

But sometimes you may face similar problem but different solution.

You are able to browse the file by CLI or GUI, but cannot delete by either way. It returns you device or resource busy or similar error messages.

You could try following command to delete the file/folders:

rm [File or folder name] -rf

Virtual Machine Disappeared on vCenter Inventory

Just googled this issue, some of people got similar problem, following was my problem and solution, hope it’s useful for you.

Virtual machine maybe lost on vCenter Server inventory for some reason, it’s also disappeared on ESXi inventory when you connect to the individual host directly. You are able to find the VM process by esxcli vm process list, but not able to get it by vim-cmd vmsvc/getallvms.

If you try load the VM manually by commeand vim-cmd vmsvc/reload [vmid], it back you error:

(vim.fault.NotFound) {
dynamicType = <unset>,
faultCause = (vmodl.MethodFault) null,
msg = “Unable to find a VM corresponding to “vmid””,
}

You can also find the VM process by esxtop command.

My solution was remove the ESXi host from vCenter Server, then restart management services, and then browse the VM folder, add the VM back to inventory, then join the host back to vCenter Server.

How to get HBA WWPN of ESXi hosts

It’s busy month, I haven’t update my blog since I back from Phuket with my wife. I’m running into multiple projects, a little overload.

Just a quick share, my storage team ask me provide WWPN of all hosts to do a health check. it’s nightmare to pull out the data from vSphere client or web client. Just found a way to get it.

Get-VMHost -Location | Get-VMHostHBA -type fibrechannel | select VMHost,Device,@{N=”WWPN”;E={“{0:X}” -f $_.PortWorldWideName}}

Especially “{0:X}” -f $_.PortWorldWideName}

{0:X} is format, check out here to find more.

-f is kind of pipeline.

$_.PortWorldWideName is the value you want to convert.