Oke, let me first start of a special thanks to my colleague’s Hans Vredevoort, Darryl van der Pijl and Mark Scholman for leading me to the solution!
Since the release of Azure Stack TP2 i did not found the time to start a deployment and play around with Azure Stack TP2. Tp1 was unstable and had a lot of strange issue’s were TP2 was much more stable I was told.. But for the past view weeks I have ben struggling with the deployment!
I’ve deployed it over and over and it keept failing on error in step 60.140.145 were the VM’s were registered at the Compute Controller.
2016-10-24 04:38:18 Error 1> 2> Task: Invocation of interface 'Migrate' of role 'Cloud\Fabric\VirtualMachines' failed: Function 'Add-CpiObjects' in module 'Roles\VirtualMachine\VirtualMachine.psd1' raised an exception: Failed to register virtual machine 'MAS-NC01' with compute controller. Operation failed with: Exception calling "GetResult" with "0" argument(s): "MakeReservation VM:36fae812-1703-4198-b0ae-39075e77e2ef Cluster:S-Cluster" At C:\CloudDeployment\Roles\VirtualMachine\VirtualMachine.psm1:2118. + $eceVm.Name at Trace-Error, C:\CloudDeployment\Common\Tracer.psm1: line 52 at Add-CpiObjects, C:\CloudDeployment\Roles\VirtualMachine\VirtualMachine.psm1: line 2118 at <ScriptBlock>, <No file>: line 18
Since I was on a VLAN tagged network and had no DHCP I was suspecting that one of these parameters –PublicVLan -NatIPv4Subnet -NatIPv4Address -NatIPv4DefaultGateway were giving me a hard time deploying TP2. But after adding a DHCP and change my VLAN to native/untagged so I wouldn’t need the parameters I still ended up at the same error!
Well, it turns out that this could have something to do with the time zone. My test servers are in the US in pacific time zone UTC-8. I am located in the Netherlands UTC+1. My server BIOS and OS time zone setting was on UTC-8. But as I stated at the beginnen of my blog a suggestion from my colleague’s was to change the timezone to UTC, which didn’t made any sense at first. So after changing the Server BIOS to UTC and prepping the Cloudbuilder.vhdx and rebooting in to Windows Server 2016 TP2 VHDX like described here, and also changed timezone to UTC I fired up the deployment without additional VLAN or NAT Parameters.
cd C:\CloudDeployment\Configuration $adminpass = ConvertTo-SecureString "a-password" -AsPlainText -Force $aadpass = ConvertTo-SecureString "a-password" -AsPlainText -Force $aadcred = New-Object System.Management.Automation.PSCredential ("username@xxxx.onmicrosoft.com", $aadpass) .\InstallAzureStackPOC.ps1 -AdminPassword $adminpass -AADAdminCredential $aadcred
After some hours of waiting I reconnected to the server and noticed a powershell window that was finished processing and only had yellow and green lines instead or RED error lines 😀
Do not yet fully understand why this is an issue but it got me past bloody step 60.140.145 🙂 . After thinking about previous deployments I did notice some VM’s not always had the correct timezone/time settings so that might be something that could trigger issue’s.
If you have any info of feedback please feel free to leave a comment or contact me on twitter or the contact page
Good luck deploying!
Regards
Pascal Slijkerman
Hi Pascal!
I have similar problem. Screenshot:
https://1drv.ms/i/s!AommlTNrLr3Gis9k7Kd9s3wa-jKBPg
That Hyper-V VM.
Do you have any idea?
Thank you!
Hi Artem,
I see, it did register almost al the machines before it failed though. Did you also check the Server BIOS? Is It also on UTC?