OM 2012 R2 – Remote SQL Instance Ports

When Installing OM 2012 R2 in a secure environment it is important to know the right ports to open, especially when you’ve gone to the trouble of setting the SQL instance port.

When installing OM, you will need to specify the SQL instance and port. What isn’t clear is that you actually need the RPC high end ports open in order for it progress past this step. Therefore the required ports are:

RPC endpoint mapper | Port number 135 | Protocol TCP

RPC/DCOM High Ports (2008+ OS) | Ports 49152-65535 | Protocol TCP

I believe the RPC ports can be closed after installation is complete.

For further details on other ports in a secure environment see Mike’s post here: http://www.systemcentercentral.com/scom-2012-ports-restricted-environment-open-ports-really-need/

VMM 2012 R2 Add Virtual Machine host failed

Trying to add an existing clustered host to VMM fails giving the following error:

There were no computers discovered based on your inputs. Please verify the following:
– Your Virtual Machine Manager Server service account has at least read access to the Active Directory domain you requested to discover
– Your Virtual Machine Manager Server has access to the computers in the discovery scope
– The servers you specified are powered on and running
– You specified valid crediantials to access the servers
– The user account that you specified has Administrator privileges on the Windows servers to be discovered
– The servers you specified are already managed by Virtual Mahcine Manager

 

Several things can cause VMM to have problems pushing out agents, the first place to always check is any firewall (software or hardware) between the VMM server and the Hyper-V host.

The following two sites can help with initial troubleshooting:

http://edhammond.blogspot.co.nz/2014/01/solving-winrm-host-connection-errors-in.html

http://support.microsoft.com/kb/2742246

 

In my case, nothing was working. I was seeing some strange behaviour from the cluster however (due to a failed orphaned resource – see my earlier post). 

I now believe the failed orphaned resource was possibly preventing the VMM agent from being deployed, when you try deploy an agent to a clustered host it detects that it is part of a cluster.

My workaround (just to get the agents installed first) was to evict the cluster node, this would allow the VMM agent to be pushed out. I then joined the node back in the cluster and failed the roles over, evicted the other node and again pushed the agent out. This allowed me to see the cluster in VMM, problems did remain due to the failed resource, but this was fixed by removing it with powershell.

VMM 2012 R2 Error 12711 (VMM cannot complete WMI operation on the server) Removing an Orphaned Clustered Resource (Failover Cluster 2012 R2)

Sometimes admins attempt to repair VMs, not knowing about the relationship between VMM/Failover-Clustering/Hyper-v they fix up a VM to a working state, leaving empty orphaned cluster resources behind, and also causing VMM to report error 12711.

To fix these resources we need to get into powershell and connect to the cluster.

Start with a “Get-ClusterResource” and see if you can see it, try remove it using the “Remove-ClusterResouce” command first, if this fails (as it sometimes does if its an empty VM resource) then we can try removing its group with contents

do a “Get-ClusterGroup” and identify the name, now try “Get-ClusterGroup nameofGroup | Remove-ClusterGroup”

still not working? -group is not empty error? thats fine try “Get-ClusterGroup nameofGroup | Remove-ClusterGroup -RemoveResouces”