This solution have been tested with the RC Release – it could be changed in the Final Release.

Last week I had an interesting case at a customer – they have a need of making the OpsMgr 2012 RC installation completely Fault tolerance – both the Operator console, Web console and report service need to be able to switch between two OpsMgr Management servers.

By default if your System Center 2012 – Operations Manager Operations console is connected to the Data Access service of one management server and the connection fails, it does not automatically failover to another management server in your management group. To enable high availability for your Operations Manager Data Access service, you can use Network Load Balancing (NLB). By configuring NLB, and connecting to the NLB cluster, if the connection fails, it will automatically be redirected to another management server.

High level: We need to do this in three stepsFirst we install NLB on the two Management Servers, next step is to install the Web console and Report Service. And we end up by configuring the services to use the NLB record created.

Low level configuration – For this sample we use the following machines:
CTMS12012 Management server #1
CTMS22012 Management server #2
CTSQL Database Server
CTIIS Internet Information Server
CTDC Domain Controller (DNS)
OpsMgrConsole.hq.com DNS FQDN for the NLB Cluster Name (Register in DNS)

Install and configure NLB

  1. Log on to CTMS12012 with a user who have permissions as the local .\Administrators

  2. Click Start and right-click Command Prompt, choose “Run As Administrator”, If you get an message box from UAC choose Yes.
  3. Type the following to install NLB: servermanagercmd.exe -install nlb
  4. Install NLB on CTMS22012 by doing pkt, 1. to 4.
  5. When CTMS12012 is ready, log on with a user who have administrative permissions
  6. Click on start, click All Programs, Click on Administrative Tools, Right Click on “Network Load Balancing Manager” and click “Run as Administrator
  7. Rightclick on “Network Load Balancing Clusters
  8. test
  9. Choose New Cluster
  10. Enter CTMS12012 and click Connect
  11. Click Next
  12. Leave Default settings and click Next
  13. In the “New Cluster window” Click Add and enter 10.1.1.100 as IP addr and 255.255.255.0 as the subnet, click OK.
  14. 2
  15. Click Next
  16. Type OpsMgrConsole.hq.com as the full internet name and choose Multicast as Cluster operation mode.
  17. 3
  18. Leave the default port rule, and click Finish.
  19. When CTMS12012 have turned green, Right click on OpsMgrConsole.hq.com and click “Add Host To Cluster
  20. Type CTMS22012 and click Connect
  21. Click Next
  22. Leave the default settings and click Next
  23. Click Finish
  24. 4
  25. In CTDC, create a DNS Host named OpsMgrConsole with 10.1.1.100 as the IP addr.
  26. If you get a warning about the reverse group, either create one or just click OK.
  27. 5
  28. In CTMS12012, start Operations Manager Console
  29. In the Operations Manager menu click on Tools and Click Connect
  30. Type OpsMgrConsole.hq.com, Click Connect
  31. From now on, the Operator Consoles are able to switch to another server (perhaps restarting the console) and you are able to reboot servers (almost) without any interference with Operations Consoles.

(This should be tested from a client and not a server.)

 

Making the Reporting Service fault tolerant

  1. On the server where you installed reporting services rightclick the file: installdrive:\Program Files\Microsoft SQL Server\MSRS10_50.MSSQLSERVER\Reporting Services\ReportServer\rsreportserver.config and click Open
  2. Click Select a program from a list of installed programs, click OK
  3. Choose Notepad and click OK
  4. Search (CTRL+F) for the tag/text: Servername and change it to the NLB Cluster name eg. OpsMgrConsole.hq.com
  5. Close and save the rsreportserver.config file
  6. Click Start and type Regedit
  7. Click HKLM\SOFTWARE\Microsoft\Microsoft operations Manager\3.0\Reporting
  8. Modify data for DefaultSDKServiceMachine to the NLB Cluster name eg. OpsMgrConsole.hq.com
  9. Restart the SQL Server Reporting service
  10. Try to execute the a report like the generic – Most Common Alerts report.

Making Operations Manager Web console fault tolerant

  1. On the Web console IIS Server, Open Windows Explorer and browse to:
  2. C:\Program Files\System Center Operations Manager 2012\WebConsole\WebHost
  3. Right click the file Web.config file and Open it in Notepad
  4. Find the text: managementserver name
  5. 6
  6. Change it to OpsMgrConsole.hq.com
  7. Change the autoSignOutInterval=”30” to “0” for Never sign me out.
  8. 7
  9. Save the file and run a IISRESET in cmd.exe

 

If you are presenting your data on a big screen you could also change the GlobalDashboardRefreshInterval which is 60 seconds by default – I could’t make it work with a smaller value – but a larger is ok.

Connect Powershell to your NLB

In the file c:\Program files\System Center Operations Manager 2012\Powershell\OperationsManager\StartUp.ps1, change the command, Please run a Powershell as Administrator to get permissions to write to the directory:

Start-OperationsManagerClientShell –ManagementServerName: “” –PersistentConnection: $true –Interactive: $true;

to include your NLB name like this:

Start-OperationsManagerClientShell –ManagementServerName: “OpsMgrConsole.hq.com” –PersistentConnection: $true –Interactive: $true;

MAY YOUR OpsMgr 2012 be: $true