Archive for the ‘Tips&Tricks’ Category

VMM Tips & Tricks: Add VM to User Role job fails with Error 20413

December 1, 2014 1 comment

In New VMM 2012 R2 with CU3 implementation when you try to add VM to existing user role you may see below error

Error (20413)
VMM encountered a critical exception and created an exception report at C:\ProgramData\VMMLogs\SCVMM.4e32da1c-645d-42b3-8c79-612d502ca2ff\report.txt.

Recommended Action
See the report for more details and search user forums for well-known failure root causes for self-help.

VMM Refresh Error



After some investigation I found the error related to VMM DB, After applying the CU3 you need to run same SQL script described in CU1, CU2

Important After you install the update package, you must apply the following SQL script on your Virtual Machine Manager Microsoft SQL Server database for Update Rollup 2 to function correctly.


Please make sure to have a backup from VMM DB before any modification to it. Once script run successfully the VM can be added to user role without any issue

Categories: Tips&Tricks, VMM2012

SCSM Upgrade Tricks: Part 1: Preparation tasks

December 5, 2012 Leave a comment

I will start a series for SCSM upgrade tricks that I faced during SCSM upgrade project I went through in the last 6 months.


Today we will start with Preparation tasks". First of all you can check Microsoft TechNet links

1- Upgrade to System Center 2012 – Service Manager

2-Failed Upgrade in System Center 2012 – Service Manager


1- Make Sure that Service Manager Servers “Management Server, Data warehouse Server” are updated to Service Manager 2010 SP1 CU4 (Build Number 7.0.6555.131). And make sure

2- Make sure that all Servers OS is running Windows 2008 R2 SP1.

3- Connectors – Please make sure that your connectors are NOT running while you do the upgrade, the upgrade will work, however your connectors will run and attempt to catch up with it. This will in turn, create additional load on the SQL Service on your Management Server, which will in turn affect overall performance and will slow down workflows in particular.

4- Backup all SCSM databases and the SSRS databases “Either using backup Solution or SQL backup”

5- Export and back up unsealed Management Packs since these are the ones which contain customizations and really should be backed up.

a. Create a directory e.g., C:\mpexport that will be the target directory for the export.

b. Load the Service Manager Cmdlet Snapin, so from a PowerShell session type:

c. PS> add-pssnapin SMCmdletSnapin

d. PS> Get-SCSMManagementPack

e. This should show you the list of management packs, then export them to a directory

f. PS> Get-SCSMManagementPack|where-object {! $_.Sealed}|Export-SCSMManagementPack –Directory C:\temp (The use of Where-Object filters the result set to only those management packs which are not sealed).

g. Now back up these management packs.

6- Remove unsupported components for the upgrade like “Microsoft Codeplex Incident SLA Management package”

7- Clients connecting to self-service portal: Clients will need to have Silverlight 4.0 installed and running Internet Explorer 8 or 9. Firefox is not supported as a browser. We need to make sure that all Clients have Silverlight 4.0

7- Remove unnecessary software: As a recommended practice, anything not related to system operation should be removed from servers. Also some of the software on servers may interfere with the upgrade and should be removed like Microsoft SCOM 2007, 2012.

8- Check free space on servers: There should be at least 1 GB free space on server for upgrade.

9- Close management consoles: Management consoles should be closed before starting the upgrade process.

10- Install all perquisites for SCSM 2012

11- Back up your Service Manager and data warehouse encryption keys.

12- Install SQL Server Analysis for data warehouse reporting

Categories: SCSM, Tips&Tricks

VMM Tricks: Remove old VMM server name from VMM Server navigation pane

November 27, 2012 Leave a comment

One question on VMM forums was “How to remove old vmm2012 server name from VMM Server navigation pane?”


Answer is simple: check these tables “In this case it was [tbl_VMM_Server]

check these tables





Please note that direct editing VMM database may affect VMM Server so please do that carefully

Categories: Tips&Tricks, VMM, VMM2012

VMM Tricks: VM Migration failed with Error (2912): An internal error has occurred trying to contact an agent on ServerFQDN

October 14, 2012 1 comment

Symptom :

When you try to migrate a VM between two Hyper-V hosts using VMM, it fails with

“Error (2912): An internal error has occurred trying to contact an agent on the Server FQDN. (Element not found (0x80070490))”.



“0x80070490” means “Element not found” and “Certificate not found”. This issue may be caused by a problem with the Host certificate (incorrect name, IP instead of FQDN or NetBIOS) or the certificate is missing from the VMM server.


The possible solution is to remove the managed host from the VMM server and also delete any residual certificates from the host on the VMM server, and then re-add the host. use the following steps:

1. On the SCVMM server, remove the managed host from the console. The steps on how to remove a managed host are outlined in the following TechNet article:

2. Locate, backup, and delete any certificates for the Host computer.

A. Open the Certificate console on the SCVMM server.

  • Open a new mmc and add the certificates snap-in.
  • Select the option of ‘computer account’ and ‘local computer’.
  • Select Finish and Ok to load the snap-in.

B. The certificates for the Host computer can be in any of the following locations.

I. Personal Certificates.

II. Trusted People (if the host is W2K8).

III. Trusted Root Authorities (If the host is W2K3).

C. In each store, expand the Friendly Name field and locate the certificate[s] for the Host server that have a Friendly Name starting with ‘SCVMM_CERTIFICATE_KEY_CONTAINER’ followed by either the FQDN / IP address / NetBIOS name of the Host server and delete them.

3. Re-add the host in SCVMM which recreates the certificates as needed.

For more information about this solution, please refer:

Error 2912 when a Virtual Machine Manager 2008 operation fails due to a certificate issue: "Element not found (0x80070490)"

Categories: Tips&Tricks, VMM, VMM2012 Tags:

VMM tricks : Delete VM from VMware cause “Invalid datastore path”

October 5, 2012 Leave a comment

System Center 2012 – Virtual Machine Manager (VMM) enables you to deploy and manage virtual machines and services across multiple hypervisor platforms, including VMware ESX and ESXi hosts. In VMM, support for ESX is optimized for virtual machine and service management. VMM enables you to manage and provide resources from multiple hypervisors and make the resources available to private cloud deployments, all from a common user interface and common command-line interface (CLI).

VMM 2012 helps customers to migrate from VMware to Hyper-v using V2V feature. I saw a common case that happened for 2 of my customers

Symptons :

When you try to delete VM that are hosted on VMware Cluster from VMM Console or Sync VMM with vCenter you get this error.


Error (2947)
Virtual Machine Manager cannot complete the VirtualCenter action on the server XYZ.Domin.Local because of the following error: Invalid datastore path ‘[ESX_Datastore] ‘
Unknown error (0x194)

Recommended Action
Resolve the issue and then try the operation again.



This error happens after migration of VM from VMware to Hyper-V using VMM. if you delete the “VM files, Storage” from VMware cluster using the vCenter console “not the VMM Console “ and then try to Refresh VMware Cluster you still able to see the VM. If you tried to delete it again from VMM Console you will get this error


You can use PowerShell to force remove that VM:

$VM = Get-VM -Name "VM Name"
Remove-VM -VM $VM –Force

Categories: Tips&Tricks, VMM, VMM2012, vmware Tags:

VMM tricks: VMM 2012 Cluster, Error 441 Agent Communication is blocked

We have 2 Cluster nodes VMM 2012 , VMM 01 and VMM 02 and now we are able to add the Hyper-v cluster after solving this problem.

We are able to discover all VM and everything looks fine, After 2 days we found a flood of errors in VMM console related to VMs


After some troubleshooting we found the problem, In VMM 02 the WinRM service was not configured. Simple run command WinRM Quickconfig and all errors disappear

Categories: Tips&Tricks, VMM, VMM2012

VMM Tricks: VMM 2012 error when adding new Host

June 15, 2012 2 comments

We are working on new private cloud implementation with high restrictive security boundaries, while adding new Hyper-v cluster “ Windows 2008 R2 Sp1 Core edition” we got error



We make sure that we are using administrator account, No firewall problems.

We have Parent Domain (xxx.local) where all Management servers are running "VMM, SCOM and DPM" and we have child domain ""  where all Hyper-v servers are running "10 Servers in One Hyper-v cluster using Windows 2008 R2 SP1 Server Core"
the Hyper-v hosts are from HV01 to HV10 and using IPs from XX.XXX.XX.51 to XX.XXX.XX.60 and Hyper-v cluster IP is XX.XXX.XX.61


When I am trying to ping any node of the Hyper-v hosts it is working fine, Able to resolve the Hosts names using NSLOOKUP

When I am trying to add any host to Hyper-v cluster I got this error
[2124] 084C.06D4::06/03-09:25:30.427#18:ServerConnection.cs(1229): Microsoft.VirtualManager.Utils.CarmineException: HV01 cannot resolve with DNS.

I tried to add the host name "netbios and FQDN" to hosts file and got the same error. the funny thing that when I am trying for example HV04 I got the same error that the servers HV01 cannot resolve with DNS.  !!!!!!!!!!

VMM is trying to resolve HV01 not Hv04, I tried IPs and got the same error.

After some troubleshooting it turn to be something wrong in the VMM Server

Using Network Monitor and Wirshark here what i found

When adding the cluster IP or FQDN , VMM Service check with Parent DC to resolve and it manage to get the IP "it does not matter wherever I provide IP or FQDN"

Then the VMM checks with the Child DC for the cluster "Since The cluster and Hyper-v hosts exist in the child domain" and it resolve the cluster name.

VMM start to query the cluster nodes and resolve all nodes and suddenly it start searching for any random host NetBIOS name and consider Host.Parent.domain not hots.child.parent.domain without any reasons and return error that it can not resolve the name

So for the seek of troubleshooting we added all the hosts NetBIOS and FQDN to hosts file and got the same error. Analyzing the capture I found that all Cluster name and all hosts were resolved but suddenly the VMM did the same again with Cluster name "Since it is not added to hosts file"

Adding all Hosts and Cluster names to Hosts file solved the problem.

%d bloggers like this: