Says Tim: "You might have noticed that the SCCM Reporting Dashboard (Beta) was released about a month ago.  But what most of you don’t know is that you can use the dashboard for SCOM data.  The cool thing with the SCCM dashboard is that it’s and OPEN platform.  You can feed it a SQL query and it will display a live gauge or a graph.  This can be used query data from  the SCOM Operations DB, the Data Warehouse DB or any DB you want."

Read the entire tutorial (at the source)

Check out Ake’s article at: http://blogs.technet.com/momteam/archive/2010/03/15/help-us-help-you-operational-data-reporting.aspx

Very cool to see someone forging ahead with OpsMgr X-Plat authoring....Go Kris!

Says Kris: "Over a series of future posts, I intend to describe my efforts to build an OpsMgr SCX Management Pack for monitoring of Oracle running on UNIX/Linux platforms.   In this initial post, I will describe the first step in building this MP:  creating a custom shell command invoker probe action.  As I had described in this previous post, the Microsoft.Unix.WSMan.Invoke.Probe in the Microsoft Unix Library MP provides a wrapped version of the Microsoft.SystemCenter.WSManagement.Invoker.   However, I wanted a bit more granular control of the invoke action (e.g. modifying the command timeout) and felt it appropriate to create a custom Probe Action to wrap the Microsoft.SystemCenter.WSManagement.Invoker module and decided to create a similar, but custom probe action"...

Follow Kris' efforts on his Operating Quadrant blog:

Building an Oracle Management Pack for OpsMgr Cross-Platform Agents, Part 1

This query is computers with Office 2007 Professional installed

elect SMS_R_SYSTEM.ResourceID,SMS_R_SYSTEM.ResourceType,SMS_R_SYSTEM.Name,SMS_R_SYSTEM.SMSUniqueIdentifier,SMS_R_SYSTEM.ResourceDomainORWorkgroup,SMS_R_SYSTEM.Client 
from SMS_R_System where SMS_R_System.ResourceId in (select distinct SMS_G_System_ADD_REMOVE_PROGRAMS.ResourceID   
from SMS_G_System_ADD_REMOVE_PROGRAMS   where SMS_G_System_ADD_REMOVE_PROGRAMS.DisplayName = 'Microsoft Office Professional 2007')

And the NOT version. This query is computers without Office 2007 Professional installed

elect SMS_R_SYSTEM.ResourceID,SMS_R_SYSTEM.ResourceType,SMS_R_SYSTEM.Name,SMS_R_SYSTEM.SMSUniqueIdentifier,SMS_R_SYSTEM.ResourceDomainORWorkgroup,SMS_R_SYSTEM.Client 
from SMS_R_System where SMS_R_System.ResourceId NOT in (select distinct SMS_G_System_ADD_REMOVE_PROGRAMS.ResourceID   
from SMS_G_System_ADD_REMOVE_PROGRAMS   where SMS_G_System_ADD_REMOVE_PROGRAMS.DisplayName = 'Microsoft Office Professional 2007')

How to create custom collection with a query

Here's a write up on how to add your query to a custom collection -  ConfigMgr: How to Create a Dynamic Collection with a Query

select SMS_R_SYSTEM.ResourceID,SMS_R_SYSTEM.ResourceType,SMS_R_SYSTEM.Name,SMS_R_SYSTEM.SMSUniqueIdentifier,
SMS_R_SYSTEM.ResourceDomainORWorkgroup,SMS_R_SYSTEM.Client from SMS_R_System 
inner join SMS_G_System_COMPUTER_SYSTEM on SMS_G_System_COMPUTER_SYSTEM.ResourceId = SMS_R_System.ResourceId 
where SMS_R_System.OperatingSystemNameandVersion like "%Workstation 6.1%" or SMS_R_System.OperatingSystemNameandVersion like "%Windows 7%

How to create custom collection with a query

Here's a write up on how to add your query to a custom collection -  ConfigMgr: How to Create a Dynamic Collection with a Query

select SMS_R_System.Name, SMS_R_System.SMSAssignedSites, SMS_R_System.IPAddresses, SMS_R_System.IPSubnets, 
SMS_R_System.OperatingSystemNameandVersion, SMS_R_System.ResourceDomainOrWorkgroup, SMS_R_System.LastLogonUserDomain, 
SMS_R_System.LastLogonUserName, SMS_R_System.SMSUniqueIdentifier, SMS_R_System.ResourceId, SMS_R_System.NetbiosName 
from SMS_R_System where SMS_R_System.OperatingSystemNameandVersion like "%Server 6.0%"

How to create custom collection with a query

Here's a write up on how to add your query to a custom collection -  ConfigMgr: How to Create a Dynamic Collection with a Query

Here's another I've been playing with. ConfigMgr collection query with variables. Here's a collection query that

SELECT * FROM SMS_R_System
WHERE SystemOUName like "%PLAY%

How to create custom collection with a query

Here's a write up on how to add your query to a custom collection -  ConfigMgr: How to Create a Dynamic Collection with a Query

Here's a ConfigMgr 2007 collection query I've been using to find Systems without an OpsMgr 2007 agent installed. 

This query will show you systems with the System Center Management service (HealthService) installed

select SMS_R_SYSTEM.ResourceID,SMS_R_SYSTEM.ResourceType,SMS_R_SYSTEM.Name,SMS_R_SYSTEM.SMSUniqueIdentifier,
SMS_R_SYSTEM.ResourceDomainORWorkgroup,SMS_R_SYSTEM.Client 
from SMS_R_System where SMS_R_System.ResourceId in (select SMS_R_System.ResourceId from SMS_R_System 
inner join SMS_G_System_SERVICE on SMS_G_System_SERVICE.ResourceID = SMS_R_System.ResourceId 
where SMS_G_System_SERVICE.Name = "HealthService")

And this one will show you all systems without that service

select SMS_R_SYSTEM.ResourceID,SMS_R_SYSTEM.ResourceType,SMS_R_SYSTEM.Name,SMS_R_SYSTEM.SMSUniqueIdentifier,
SMS_R_SYSTEM.ResourceDomainORWorkgroup,SMS_R_SYSTEM.Client 
from SMS_R_System where SMS_R_System.ResourceId NOT in (select SMS_R_System.ResourceId from SMS_R_System 
inner join SMS_G_System_SERVICE on SMS_G_System_SERVICE.ResourceID = SMS_R_System.ResourceId 
where SMS_G_System_SERVICE.Name = "HealthService")

How to create custom collection with a query

Tommy did a nice write up on this at ConfigMgr: How to Create a Dynamic Collection with a Query

Problems with the Health Service Watcher (Kevin)

select bme.DisplayName, s.HealthState, m.MonitorName, dateadd(hh,-5,s.LastModified) as 'LastModifiedCST (GMT-5)'
from BaseManagedEntity bme
inner join state s on s.BaseManagedEntityId = bme.BaseManagedEntityId
inner join Monitor m on m.MonitorId = s.MonitorId
where HealthState in (2,3)
AND MonitorName = 'Microsoft.SystemCenter.HealthServiceWatcher.HealthServiceAvailabilityState'
AND bme.IsDeleted = 0
order by bme.DisplayName, m.MonitorName

Monitor for the Health Service Watcher class with issues (Kevin)

select bme.DisplayName, s.HealthState, m.MonitorName, dateadd(hh,-5,s.LastModified) as 'LastModifiedCST (GMT-5)'
from BaseManagedEntity bme
inner join state s on s.BaseManagedEntityId = bme.BaseManagedEntityId
inner join Monitor m on m.MonitorId = s.MonitorId
where HealthState in (2,3)
AND MonitorName like 'Microsoft.SystemCenter.HealthService%'
AND bme.IsDeleted = 0
AND m.IsDependencyMonitor = 0
order by bme.DisplayName, m.MonitorName

Agents not reporting (grey), Not Pingable, etc. (no idea where this came from – Pete maybe)

Here is the 'Agents Not Pingable' query that I use

SELECT bme.DisplayName,s.LastModified as LastModifiedUTC, dateadd(hh,-5,s.LastModified) as 'LastModifiedCST (GMT-5)' 
FROM state AS s, BaseManagedEntity AS bme 
WHERE s.basemanagedentityid = bme.basemanagedentityid 
AND s.monitorid 
IN (SELECT MonitorId FROM Monitor WHERE MonitorName = 'Microsoft.SystemCenter.HealthService.ComputerDown') 
AND s.Healthstate = '3' AND bme.IsDeleted = '0' 
ORDER BY s.Lastmodified DESC

And you can actually find grey agents in general using PowerShell

WCC = get-monitoringclass -name "Microsoft.SystemCenter.Agent"

$MO = Get-MonitoringObject -monitoringclass:$WCC | where {$_.IsAvailable -eq $false} 

$MO | select DisplayName

Or you can use SQL

SELECT ManagedEntityGenericView.DisplayName, ManagedEntityGenericView.AvailabilityLastModified

FROM ManagedEntityGenericView

INNER JOIN ManagedTypeView ON ManagedEntityGenericView.MonitoringClassId = ManagedTypeView.Id

WHERE (ManagedTypeView.Name = 'microsoft.systemCenter.agent') AND (ManagedEntityGenericView.IsAvailable = 0)

ORDER BY ManagedEntityGenericView.DisplayName

More OpsMgr SQL Queries

Kevin has a list HERE

Pete has some cool queries in this Database Hygiene Series

Although the approaches may be the same for most steps in Visual Studio 2008 the reports created in Visual Studio are NOT backwards compatible with Sql Reporting Services 2005. Reports designed with visual studio 2008 can only be used if you are running SQL 2008 reporting services! So for SQL reporting 2005 you should visual studio 2005 instead!

Wouldn’t it be nice to design your own reports with nice looking bars or graphs with your company logo on it ?

Requirements before starting are:

• Visual Studio 2008 which is delivered with SQL 2008
• Authoring Console
• Feeling really artistic!

Before we start with opening the Visual Studio console you will first need to determine which counters you want to generate the report on.

The example will create a custom report for reporting on logicaldisk space. Since this is one of the most asked for and is missing from the default reports.

Counter(s) of interest

We are going to start with % Free Space.

This is the information we need to know to be able to collect the counters:

Counter : % Free Space

Object: LogicalDisk

The other ones are also of interest but to start with we need to know the object and counter to be able to setup a dataset to retrieve the information from the Datawarehouse.

The begin

Open visual studio and follow the steps below to start a new project in which we are going to create the report for disk space.

Go to new and start a new project. Select a Report Server Project and give it a Name.		After Creating the project we are going to create a report within this project. Navigate to Solution Explorer on the right side of the screen. (if not shown go to view and select Solution Explorer) Right Click on Reports and select Add New Item…
Select Report and give it a Friendly Name and click Add.		Now we have a report in our project. Next we are going to add an Item to the report. As stated in the design surface you can add one by using the toolbox on the right or you can right click and select insert.
When you right click and select insert you have the following options you can select. For our example we are going to select a Matrix.		When the Matrix is selected we need to define a dataset. This is going to be the query we are going to use to query the OperationsManagerDW. Give it a friendly name and select New next to Datasource.
The Datasource is the Shared datasource the Reporting server uses to connect to the OperationsManagerDW. The Name is “Data Warehouse Main” Unfortunately Visual studio 2008 doesn’t support datasource names with spaces… so we fill in “DataWarehouseMain” instead. We change this afterwards. Next connection string, simply edit browse your sql server for the OperationsManagerDW. The connection string to you data warehouse is used to test our reports directly from visual studio.		After Setting up the datasource we can add the query we are going to use. For ease click Query Designer… The Query Designer will open and you need to add your query in the box with SELECT FROM

Main Dataset Query

This is the query we are going to use. The SELECT and FROM part are for collecting the data from the Perf.vPerfDaily view which is the view for Daily data from the OperationsmanagerDW.

TIP! If you change vPerfDaily to vPerfHourly you will retrieve the data per hour!

The WHERE part is the part we are filtering the data on. Notice we are going to filter on date @Start_Date and @End_Date which will be the parameters we are going to define. This way we can tell from which date to which date we want to run the report on.

The data is filtered on LogicalDisk which will only retrieve Data from the Object Logicaldisk.

Last we are also filtering on @ServerName which will also be a Parameter we are going to declare for the servername we want to retrieve the data.

Lastly the data is ORDER BY date. This way the data is retrieved in a chronological way.

Well after pasting in the query you can run it by pressing “!”. It will ask for the parameters fill in a start date / end date and server name (FQDN). The query designer now shows the data and how it collected the data by joining the tables. How cool is that and you are starting to look like a developer now already with this cool code and models ;-) After looking over your result. You will notice we are retrieving more data then we are actually going to use in this example. No worries you will probably have enough inspiration after designing the report to use the other data as well! When you are done looking at your developer like screen you can click ok and save your dataset settings.		We are going to create 1 more dataset and after that declare the parameters. First the Dataset, let’s call it DataSet_Servers Use the following query: Select DISTINCT vManagedEntity.Path FROM Perf.vPerfDaily AS vPerf INNER JOIN vPerformanceRuleInstance ON vPerformanceRuleInstance.PerformanceRuleInstanceRowId = vPerf.PerformanceRuleInstanceRowId INNER JOIN vManagedEntity ON vPerf.ManagedEntityRowId = vManagedEntity.ManagedEntityRowId INNER JOIN vPerformanceRule ON vPerformanceRuleInstance.RuleRowId = vPerformanceRule.RuleRowId Where ObjectName = 'logicaldisk' This query retrieves all Server Objects to report on.
Declaring the parameters in the last thing before we can start with designing the report. These are our parameters: @Start_Date @End_Date @ServerName Go to parameters and go their properties.		At the parameters properties for @End_Date set the data type to Date/Time on the general tab. Next Default Values tab and select Specify values. This will set the default values for the parameters when opening the report. For Value click the Function button (fx) and now you are in the expression designer. Go to Common Functions\Date & Time\ and select today (double click). TIP!Also there are examples how to use this function.You can select each to find what it is used for! Function for End_Date which should be today =Today() Do the same for @Start_Date with function =DateAdd("d",-7,Today()) Finally the @Servers at the General Tab select Multiple values.Next open the Available values tab and select Get Values from Query. Fill the Dataset we created DataSet_Servers and fill in both Value field and Label field with Path. Simple use the selection to select the dataset and the values.
Now we can start with the design drag and drop the values from the report data pane onto the Matrix like on the right. You can now run a test report to check which results you get.		When you did test the report you noticed there where many counters and data. First we going to filter the results since we are interested in the % Free Space for this example. To filter the results we need to change the Column Group Countername. Open the group selecting the column and go to the group properties. Now navigate to Filters. Here we are going to filter our results. Click Add to add a filter and in expression select [CounterName] use the = sign as Operator and fill in % Free Space. Select Ok and run the report again to check the results. running the report will tell you how the data is presented and give you a quick solution to be able to visualize what you are actually doing. After Adding the filter we have narrowed the results down to only % Free Space. Except now the result is calculated all results together.
We need to change the results to 1 value which makes sense and probably the most up-to-date one. Select the cell and right click to open the FX Expression designer again. Almost every value in the report has a possibility to add expressions! This makes it very flexible! Now change the expression to the following: =Round(Last(Fields!AverageValue.Value)) The Round function is used to round the number so it’s easier to read instead of a PI like number ;-) The Last function is used to only retrieve the last value. Which is today because we ordered the query on Date we now this is the last number!		The result should now look something like the right picture. By adding the filter on the Countername we only see the counter % Free Space. And by adding the expression in the details part =Round(Last(Fields!AverageValue.Value)) we have the results shown. The screen is just from a test environment and the results are probably different but the basics should look the same. Computer \ Instances running on the computer and free space. The next steps will draw the graph in a more nicer way.

The Designing Part

The designing part really depends on your personal favor but I will show you how you can make your report look more sharp in a couple of minutes.

Let’s add some more to the report. Open the toolbox and drag and drop a Gauge onto the drawing area. Wow aren’t those meters looking sharp! Select the one you want to use and ok. The meter will be pasted into your design as a new object. Simply drag and drop your new meter object into your matrix and let it go in the cell with =Round(Last(Fields!AverageValue.Value)) .		First let’s go to report properties to check and maybe change the page settings of the report. The values given here you should pay close attention since when designing you should always make sure you stay within these boundaries. Especially when you want the report to be converted to like.pdf. When you go over the settings the graph will expand multiple pages which isn’t always nice looking. If you look at the default values here you will notice the width is 8.5 and margins both right and left 1 in this makes your drawing area 6.5 same goes for the height of your page which would be 9. Make sure your designing area is within 6.5 and 9! Simply click on your designing area and go to it’s properties if no properties are shown on the left of the screen click F4 to make them appear. Now change your design area to 6.5 by 9. This way you are always sure you are editing within the page limit! TIP! When you get blank pages it is most of the times because these boundaries are not set correctly!
The matrix should now look something similar. you can expand the matrix now from one point of your designing area to the other and make the gauge more visible. You can select the different objects which make up the gauge graph. The important one is the bar as shown. For both pointers add the expression =Round(Last(Fields!AverageValue.Value)) This will show the %free space collected value but now on the bar!! TIP! when designing the bar first make it look like you want and after this you can edit the size. This makes navigation easier ;-)		After playing around endlessly with colors and settings let’s continue we have a report to make  ;-) Let’s make a page header and footer to hold extra information like title company logo execution time etc. Right click on your design area to add a page header and footer.Here you can add a text box or image. In the text box you can also add expressions! Or drag and drop report properties to the text box. This way you can display logos, a title, creation date, etc.of the report. Save your report as .rdl file and open the SCOM Authoring Console.
Create a new Management pack and go to the reporting tap. Create a new report and give it a name. Next go to the Definition tab and select “Load content from file” Navigate to the .rdl file and import it.		Now you will see the xml version of your report in the authoring console. First remove the first line After removing this line we need to correct the datasource setting. Remove the connection settings Data Source=.;Initial Catalog=OperationsManagerDW         true       And add after the following Data Warehouse Main. TIP! Don’t forget to change the options tab the visible setting to True otherwise no report will be shown!!! Now you can save your report in the management pack. And after this you can import the management pack in your environment.

Result

This post described how to create a report for SQL 2008 reporting with visual studio 2008. Although it takes some time to get the graphical results you want you can now create a custom report.

This report is a simple example of how to create reports using visual studio 2008. I know there are more and possible better solutions and queries to get results but this is for another post ;-)

Have fun with expanding your reporting functionalities and drawing skills!

To pause script execution for 30 seconds in Windows 2003 and older was easy enough with the script command

sleep 30 

on Windows 2008 and Windows 7, you can use this alternative in your batch files

@ping -30 127.0.0.1 > nul

All in all, it makes me glad we have PowerShell. Since I had the script I'd carried forward from Windows 2003 already written (that used Sleep and EventCreate to trigger a unit monitor in OpsMgr), I didn't want to start from scratch.  But the PowerShell equivalent is simple as well...I just have to convert the rest of the script when I have some free time.

Sleep in seconds

Start-Sleep –s  30

And now, back to work....

Pete Zerger, MVP OpsMgr and Essentials

]]> Sat, 13 Mar 2010 06:12:03 GMT /BlogDetails/tabid/143/IndexId/60799/Default.aspx This particular issue has been one of the banes of my existence since I started deploying large-scale Operations Manager 2007 environments. I follow the best instructions available to the letter and create my certificates, use the MOMCertImport tool, verifying connectivity between systems, approve and install the gateway and… no-go. My gateway server shows up in the administrator pane, management servers and is stuck forever in ‘Not Monitored’. Several years back a co-worker of mine coined a term which keeps coming back to mind when this happens – the term is ClusterTruck. He coined this term because the original word kept getting blocked by the email spam filters (I’ll refrain from adding it to this blog but I bet you can guess the original).

I reached out to my favorite PKI geek (JC) and he helped to verify that the certificates looked good. His assessment of the situation was specific to me, as I am the only OpsMgr administrator that he works with but his quote is right-on-mark with someone like me who is a self-professed PKI-Challenged-person “Personally, I think it revolves around SCOM administrators complete reluctance to understand the basics of PKI…..”

I worked with another one of my OpsMgr administrator co-workers (Steve) who helped to provide a second perspective and we stepped through the logs, configured additional diagnostic logging and didn’t find a specific root issue. However, this approach led to some good debug steps which may help if you hit this situation.

• OpsMgr Log on GW: In the Operations Manager log for the Gateway server, there was an informational message from the OpsMgr Connector which was a 20053 and indicated that Operations Manager had a functional certificate. I took this to mean that the issue was NOT with the certificate on the Gateway server.
• OpsMgr Log on MS: On the management server there was an informational message from the OpsMgr Connector which was a 21022 and indicated that Operations Manager had not loaded a certificate and could not communicate with systems
• OpsMgr Log on the RMS: On the RMS there was the same 21022 informational message.
• Registry on RMS/MS: There was no registry entry defined for the ChannelCertificateSerialNumber value in the HKLM\Software\Microsoft\Microsoft Operations Manager\3.0\Machine\ on either the RMS or MS servers.
• From this we summarized that there was an issue with the management servers and how their certificates had been added to OpsMgr using the MOMCertImport Utility. After digging through this blog entry http://joeelway.spaces.live.com/blog/cns!2095EAC3772C41DB!1947.entry, http://www.scom2k7.com/2009/10/ we identified that the wrong version or the wrong type (i386 vs amd64) of the MOMCertImport utility or it had been run without admin rights due to UAC (start command prompt, run-as administrator). This is also available in the Operations Manager Unleashed book and is discussed on pages 519-523. What was really interesting was that we verified that the command said that it had completed successfully.

Summary: If your new Gateway server(s) are stuck in ‘Not Monitored’ state, check the HKLM\Software\Microsoft\Microsoft Operations Manager\3.0\Machine\ChannelCertificateSerialNumber to see if it exists and has a correct value. If not, you may have run the wrong version of the utility, wrong type (i386 vs amd64) or it may not have been run-as an administrator due to UAC.

Additional: Steve Ross had written a solid document which details step-by-step on how to install and configure your gateways. This document was invaluable to getting my most recent Gateways installed and working and is available for download at http://ross-family.org/blog/2010/03/10/system-center-operations-manager-2007-r2-how-to-install-a-gateway-server/.

From the desktop to the datacenter, between the physical and the virtual.

Hi everyone, we start our first in the series on ‘Client Deployment with System Center’, and logically we begin with one of the biggest challenges every customer in the world faces.  Applications.

Organizations are becoming more and more engaged in what is running on client systems , using new tools, adding levels of management control and trying to manage new working styles and locations.  At the same time end users are placing heavier demands on the IT department to support what they need to do their jobs.  From hardware preference, to devices, to the applications that they run.  It might be as thin as a Twitter app, or as robust as a collaboration or productivity tool. 

Applications are a broad space.  There are already over 10,000 Android apps.  There are already over 25,000 Windows mobile apps and over 180,000 iPhone apps.  The Windows division has over 7000 applications in their compatibility database that are actively tracked (and referred to in the local sysmain.sdb) , with an awareness of over 1,000,000 through avenues such as Watson reporting and Online Crash Analysis.  According to market researcher DataMonitor (and linked from Wikipedia), the size of the worldwide software industry in 2008 was US$ 303.8 billion, an increase of 6.5% compared to 2007.   The Americas account for 42.6% of the global software market's value. DataMonitor forecasts that in 2013, the global software market will have a value of US$ 457 billion, an increase of 50.5% since 2008.

Source: StatCounter Global Stats - Operating System Market Share

One of the biggest challenges in the application management space is ‘what are you running’?  A memorable customer visit with a large bank still sticks to mind.  We worked closely for 6 months to define what an application was for their 180,000 users.  Their first list was 16,000 applications. When the exercise was complete, the number was closer to 2,500.  Not bad for their size, but it was all about defining what an application was. Many organizations have applications through ISVs, like Microsoft.  In addition though, there are local apps, internally written or built by contractors or vendors.  In a nutshell, ‘unclassified’ apps (we are not trying to oversimplify this).  In order to really define what an application is, its also important to consider the delivery mechanism.  All of these numbers grow exponentially when we consider the format of the app.  Full install, virtual, presentation, device or other. 

In order to understand what an application is and where it is installed, it’s important to be aware of what tools are available (eg. here) .  There are several areas of Application Management that we will detail here and in some following posts, and highlight System Center and partner solutions for you to investigate wherever we can.  Here is the topic lineup for Applications, our first part on the series:

• Inventory/Discovery – Application and Hardware/OS inventory are critical to understand before any Client deployment.  What applications are installed, what apps are actually being used, how many versions of applications are in use and where can standards and consolidation help reduce costs – are key areas for focus.  Often over the life of a client, apps get versioned, updated etc.  The OS as well goes through patch cycles, updates and Service Packs .  All of this adds up to potential variations in the landscape of clients.  To get accurate data, the right tools need to be used.  MAP, SW/HW Inventory, ACT (and it’s connector), SAM and more need to be considered.
• Compatibility  - Once your organization has the knowledge of what is out there, its time to think about what is actually moved forward (migrated).  Application consolidation, compatibility and standards with the newer OS are important to consider.  Without the right data, or tools, this becomes a massive task to get accomplished. 
• Packaging – Passing through data of inventory and compatibility takes us to the next step of making packaging decisions.  At this point decisions about what apps will be used, managed, added to SLA and help desk support lists takes us to choices about packaging, shims, formats etc.  Physical, Virtual, Presentation, Device are all formats that can be chosen – there are also others like browser based, hosted (VDI etc).  Again, tools visibility is important, so we will try here to highlight some we integrate with. 
• Deployment – Its time to deploy the client, and that means apps streamlined into the process. Chaining applications into the deployment of the OS is a strength of System Center, long known as a Software Distribution tool and from ConfigMgr an industry leader in OS Deployment.  By including applications into the OS deployment process, time and cost can be compressed significantly.  In addition, it is important to consider how to get applications distributed in production, outside of the OS deployment, for example from the help desk.  At that stage it is also important to consider Help Desk Tools integration, workflow and approvals management.  Centrally automating some of those activities from a in workflow tool is critical to retaining insight and control to the managed systems.  System Center has significant integration to this production workflow management space, and we will highlight that.
• Patch – Production management of systems must include a strategy for patch management.  Both the OS and Applications go through a lifecycle of updating, and that requires centralized strategies, tools and workflow for release.   Patching Microsoft products is only a portion of the landscape, so it is also important to highlight 3rd party support.
• Asset Management – Managing software assets through a lifecycle is equally important.  To centrally provide control and visibility to the assets installed from the perspective of installation, usage and licenses can gain an organization agility, help enforce standards, and reduce costs.  Again, the right tools, the right workflow and the right reporting are key to success.
• Reporting – By no means is this the last on the list.  Organizations today more than ever need to be dialed in on status, success, compliancy, and SLA management.  Reporting usage, workflows and compliancy are all key areas to helping IT demonstrate their abilities to be agile, responsive and a strategic asset to the organization.  System Center has made significant investments to the area of reporting through dashboards, workflow and status management.  We will highlight these for you. 

Next we will go into these with Tools analysis, interviews and recommendations for you. 

Jeff Wettlaufer 
Sr. Technical Product Manager
System Center

]]> Fri, 12 Mar 2010 03:51:19 GMT /BlogDetails/tabid/143/IndexId/60769/Default.aspx Can you write Managements Packs, have you ever thought the need to but feel you can’t? These are questions that right now are of interest to us, why?

Well if you can write an MP then you are in the minority, if you have thought the need to write one but can’t then you are in the majority! What does this mean to you……follow me and learn more..in the coming days….

Updated: May 22, 2009

Applies To: Operations Manager 2007 R2, Operations Manager 2007 SP1

The following table lists the issues that you might experience while you are using Operations Manager Reporting. It also includes the steps that you can take to resolve these issues.

Issue	Resolution
SQL Reporting Error log (Default - %Program Files%\Microsoft SQL Server\MSRS10.MSSQLSERVER\Reporting Services\LogFiles):  System.Data.SqlClient.SqlException: Could not locate entry in sysdatabases for database 'Mom'. No entry found with that name. Make sure that the name is entered correctly.	This error occurs when you install Operations Manager Reporting on a SQL Server Reporting Services (SSRS) database that is using a custom name.  To fix the problem, reinstall SQL Reporting Services by using the default names of ReportServer and ReportserverTempDB.  Then, reinstall Operations Manager Reporting.
Reporting Services Configuration Manager shows Web Service Identity as Not configured.  When you try to configure the Web Service Identity, you receive the following exception - "ReportServicesConfigUI.WMIProvider.WMIProviderException: A virtual directory must first be created before performing this operation." (SQL 2005 specific)	This error occurs when the virtual directory was not configured correctly during the SQL Reporting Services (SRS) installation.  To fix the problem, do the following: 1. In IIS Manager, delete the Reports and ReportServer Web sites under Default Web site. 2. Create new Reports and ReportServer Web sites by using Reporting Services Configuration Manager: a. In Windows, click Start, point to Microsoft SQL Server 2005, point to Configuration Tools, and then click Reporting Services Configuration. b. In the Report Server Installation Instance Selection dialog box, specify the name of the server that is hosting SQL Reporting Services, select the database instance from the list, and then click Connect. c. In Reporting Services Configuration Manager, click Server Status. Verify that Service Status is set to Running. If it is not, click Start, and then click Apply. d. Click Report Server Virtual Directory, and then click New to specify a new virtual directory. e. In the Create a New Virtual Directory dialog box, select Default Web Site from the Website list, specify a name for the virtual directory that is created by Reporting Services (or use the default ReportServer), and then click OK. f. Ensure that Apply default settings is selected, and then click Apply. g. Click Report Manager Virtual Directory, and then click New to specify a new virtual directory. h. In the Create a New Virtual Directory dialog box, select Default Web Site from the Website list, specify a name for the Report Manager virtual directory that is created by Reporting Services (or use the default Reports), and then click OK. i. Ensure that Apply default settings is selected, and then click Apply. j. Click Web Service Identity. If you do not want to change the default settings in this section, click Apply. k. Click Database Setup, and then click Connect. l. In the SQL Server Connection Dialog dialog box, specify the SQL Server name and connection details for the Reporting Services server, and then click OK. m. Click New. n. In the SQL Server Connection Dialog dialog box, specify a name for the new Reporting Services database (or use the default, ReportServer), and then click OK. o. Click Apply, and then click OK. p. Click Exit to close Reporting Services Configuration Manager. 3. Restart the IIS Admin service.
Security event logs: “ Event Source: MSSQLSERVER Login failed for user 'NT AUTHORITY\NETWORK SERVICE'. [CLIENT: ]” and “Event Source: Reporting Server (MSSQLSERVER) Report Server (MSSQLSERVER) cannot connect to the report server database.	This error occurs when the permissions for the SQL Report Server database are missing.  To fix this problem, reset the permissions by using the Reporting Services Configuration tool, as follows: 1. Open Reporting Services Configuration Manager. 2. Under Database Setup, update the Account Name to a domain account. For more information, see http://go.microsoft.com/fwlink/?LinkID=144636.
Application event log: Application pool 'ReportServer' is being automatically disabled due to a series of failures in the process(es) serving that application pool.  (SQL 2005-specific)	This error occurs when the SQL Reporting Server Service is disabled due to multiple failures.  To fix this problem, run the IISReset.exe tool, available from http://go.microsoft.com/fwlink/?LinkID=39969.
Application event log:  SQL Server Reporting Service, Service cannot be started.  The user or group name ‘LocalSystem’ is not recognized	This error occurs when the Application Pool account is switched to LocalSystem.  To fix the problem, do one of the following: · Open IIS Manager, find ReportServer or ReportServer$InstanceName under Application Pools, and then set the Identity to a domain account. · Open Reporting Services Configuration Manager. Under Web Service Identity, update the ASP.NET Service Account to a domain account.

Original post can be found here

Updated: February 17, 2010

Applies To: Operations Manager 2007 R2

The root management server (RMS) is a single point of failure for every management group. You can reduce this risk by placing the RMS role in a failover cluster that is running in quorum node configuration only. This procedure shows you how to cluster the RMS role after you have already installed the RMS role on a stand-alone server. You can place the RMS role in a cluster after the initial installation of the management group in Windows Server 2008-based clusters and in Windows Server 2003-based clusters. The procedure presented here assumes that you are using Windows Server 2008 Failover clustering. Your databases can be hosted on any of the following versions of Microsoft SQL Server:

• SQL Server 2005 with SP 1
• SQL Server 2005 with SP 2
• SQL Server 2005 with SP 3
• SQL Server 2008 with SP 1

To place the RMS role in a cluster, you have to perform the following steps:

1. Complete the pre-procedure checklist.
2. Prepare the Failover cluster nodes for the installation of the management server and RMS roles.
3. Install the Operations Manager 2007 R2 management server role on each cluster node.
4. Create the following RMS Failover cluster resources for the RMS: the System Center Management Service (HealthService), the System Center Configuration service (OMCFG), and the System Center Data Access service (OMSDK).
5. Create the virtual RMS by using the ManagementServerConfigTool.exe tool with the InstallCluster option.
6. Register the service principal name (SPN) by using the SetSPN.exe tool.
7. Verify that the clustered RMS is functioning correctly.

Supported and Unsupported Topologies

Supported Topology

This procedure works only for one starting RMS topology, which leads to a single ending RMS topology. The process is shown in figure 1. No other starting or ending RMS topologies are supported.

Figure 1: Supported topology

Three things happen in this procedure:

1. A clustered RMS server is created on a failover cluster that is not running any other workloads and that is configured as active/passive.
2. The pre-existing RMS is demoted from the RMS role and becomes a management server.
3. Any agents that had the pre-existing RMS as their primary management server are automatically redirected to the new clustered RMS. Also, all management servers are automatically redirected to the new clustered RMS.

Unsupported Topologies

Here are some examples of unsupported starting and ending topologies. If your Operations Manager 2007 R2 management group is in any of the configurations listed on the "Before" side of the diagrams, and you want to cluster the RMS, you must reconfigure the management group so that is in a supported configuration before you start the procedure.

Figure 2: Attempt to cluster the existing RMS

Figure 3: Attempt to cluster the existing RMS and to have agents report to the passive node

Figure 4: Attempt to move the RMS from one cluster to another cluster

Known Issues

• During this process, all agents that reported to the pre-existing RMS are redirected to the clustered RMS, and the pre-existing RMS (SRVR1) are demoted to be a management server (ms). If you do not want the agents to be owned by the RMS, you must manually redirect them to the demoted management server (SRVR1) or to other management servers in the management group.
• If the pre-existing RMS (SRVR1) is x64-based, and if the new clustered RMS will be x86-based, after the procedure, you have to manually demote the pre-existing RMS (SRVR1). The ManagementServerConfigTool.exe tool provides you with the command-line syntax that you must run to accomplish this.
• This procedure does not support using the Local System account for the System Center Data Access (OMSDK) or for the System Center Management Configuration (OMCFG) accounts. Domain accounts must be used and be available before you add the RMS cluster. In addition, you must ensure that the domain-based System Center Data Access account is listed as having the following fixed database roles for the OperationsManager database in SQL Server: configsvc_users, db_datareader, db_datawriter, db_ddladmins, and sdk_users.

Pre-Procedure Checklist

1. Verify that the Windows Server 2008-based failover cluster you want to use exists and is healthy. For installation and configuration guidance for Windows Server 2008-based failover clusters, see Failover Clusters (http://go.microsoft.com/fwlink/?LinkID=147285).

2. Ensure that the System Center Data Access and System Center Management Configuration accounts are configured to use a domain-based account. If you are using the Local system, change to domain-based accounts.

3. If you have to change to domain-based accounts, ensure that the System Center Data Access account is listed as having the following fixed database roles for the OperationsManager database in SQL Server: configsvc_users, db_datareader, db_datawriter, db_ddladmins, and sdk_users.

4. If you have changed to domain-based accounts, restart the System Center Data Access service, the System Center Management Configuration service, and the System Center Management service.

5. Restart the SQL Server and SQL Agent services.

6. Ensure that all System Center Operations Manager servers in the management group are running Operations Manager 2007 R2. Operations Manager 2007 SP1 and Operations Manager 2007 are not supported for this procedure. The agents in the management group can be Operations Manager 2007 SP1-based or Operations Manager 2007 R2-based agents.

7. Ensure that no agents are reporting to the management servers that are installed on either failover cluster node.

8. Disable notifications.

9. Close all instances of the Operations console that are connected to the management group.

10. On the pre-existing RMS, in the Applications and Services Logs tool, review the Operations Manager log for any error, warning, or critical events from the OpsMgr SDK Service. If you find any of these, resolve the issues before you start the RMS clustering procedure.

11. Plan an RMS network name.

12. Plan an RMS IP address.

13. Plan shared disk (clusterable disk) for the quorum drive and dedicated disk for the RMS files.

Clustering the Root Management Server in a Pre-Existing Operations Manager 2007 R2 Management Group

In this procedure, the management server and the user interface components are installed on the cluster nodes. They make use of the OperationsManager database that is hosted on an instance of SQL Server 2008.

Note

Do not install agents on the cluster nodes that support the RMS. If you want to monitor the health of a root management server on a cluster as part of another management group, use agentless management.

The following procedures show how to install an Operations Manager 2007 R2 root management server on a Windows Failover cluster. Installing Operations Manager 2007 R2 root management server on a Windows Failover cluster involves the following tasks:

• Preparing the Windows cluster and creating a Service or Application for the RMS.
• Creating Physical Disk, IP Address, and Network Name resources in the RMS cluster group.
• Installing the secondary management servers on the cluster nodes and backing up the RMS encryption key.
• Creating Generic Service cluster resources in the RMS cluster group for the RMS System Center Management service (HealthService), the RMS System Center Data Access (OMSDK), and the RMS System Center Management Configuration (OMCFG) services.
• Backing up the OperationsManager database.
• Running the SecureStorageBackup tool with the Backup option on the pre-existing RMS to back up the RMS encryption key.

  Note

  If the RMS encryption keys are successfully backed up at the end of the RMS setup, it is not necessary to back them up again at this point.

• Running the SecureStorageBackup tool with the Restore option on all secondary management servers to place the RMS key on them.
• Running the ManagementServerConfigTool tool with the InstallCluster option to instantiate the RMS cluster.
• Bringing the RMS cluster group online.
• Moving the RMS cluster group to each node to complete the configuration.
• Testing to verify successful cluster installation.
• Optionally running the ManagementServerConfigTool tool with the AddRMSNode option to add cluster nodes to the RMS cluster that were inaccessible during the InstallCluster action described previously.
• Running SetSPN.exe after the InstallCluster action.

Step 1: Prepare the cluster nodes, the RMS cluster group, and the RMS cluster group resources

To prepare the cluster nodes, the RMS cluster group, and the RMS cluster group resources

1. On each RMS cluster node, ensure that the domain Operations Manager Administrators security group has been added to the local Administrators group.

2. Ensure that each cluster node meets the prerequisites for the management server and User Interface components:

   • Windows Server 2003 SP1 or Windows Server 2008
   • MDAC version 2.80.1022.0 or later
   • .NET Framework version 2.0
   • .NET Framework version 3.0 features
   • WSMAN v 1.1 (Only required if UNIX/Linux computers will be monitored in this management group).

3. Add the System Center Data Access (OMSDK) service and System Center Management Configuration (OMCFG) service account to the Local Administrators group on each node of the RMS cluster.

4. Log on to the cluster node that will be the primary owning node for the RMS with an account that is a member of the Administrators group.

5. Start the Failover Cluster Management tool from Administrative Tools.

6. If this is the first time that the Failover Cluster Management tool has been run, you will be prompted to connect to a cluster. Select the Manage a cluster option from the Action drop-down box and either enter or browse for the cluster name for the Cluster or server name box.

7. In the Failover Cluster Management tool, right-click the Services and Applications folder, and then click Configure a Service or Application to start the High Availability Wizard.

8. On the Before You Begin page, click Next.

9. On the Select Service or Application page, select Other Server, and then click Next.

10. On the Client Access Point page, type in the network name that you have planned for your RMS. This name will be registered in DNS as an A record.

11. Click the address box and type the IPv4 address that you have planned for the RMS. This is the publicly accessible address for the RMS.

12. Click Next.

13. On the Select Storage page, select the disk resource that will be used for the RMS. This should not be the quorum disk.

14. Click Next.

15. On the Select Resource Types page, click Next.

16. On the Confirmation page, review the information, and then click Next.

17. On the Summary page, optionally review the report, and then click Finish.

18. Right-click the application that you just created, and open its properties. On the General tab, optionally select a preferred owner node and on the Failover tab, accept the default failover values and ensure that the Prevent failback option is selected.

19. Click OK.

Step 2: Check the prerequisites for the management server installation

To check the prerequisites for the management server installation

1. Log on to the cluster node that will be the primary owning node for the RMS application with an account that is a member of the Administrators group.

2. Ensure that the RMS application is owned by the node that you are logged on to. For example, in a two-node cluster, log on to node 1, open Failover Cluster Management, and, in the navigation pane, select the RMS application. The value in the Current Owner field for the RMS application should be the node that you are logged on to. If it is not, right-click the RMS application and move it to node 1.

3. On your installation media, start SetupOM.exe. This starts the System Center Operations Manager 2007 R2 Setup on the Start page.

4. Under the Prepare heading, click Check Prerequisites to start the Prerequisite Viewer.

5. In the Components box, select the Server and Console components, and then click Check.

   Note

   When these components are selected, the Prerequisite Viewer checks for the presence of Windows Server 2003 SP1, MDAC version 2.80.1022.0 or later, .NET Framework version 2.0, .NET Framework version 3.0 components, and for WSMAN v1.1.

   Note

   The results are displayed at the bottom of the Prerequisite Viewer. If there are any deficiencies, they are marked either as Warning or as Failed. Warnings can be ignored at the risk of degraded performance or other issues. Failed prerequisites must be fixed before the installation can proceed. You can close the Prerequisite Viewer, fix the items and rerun the Prerequisite Viewer checks as many times as necessary until a Passed evaluation is achieved on all items.

6. When you are done with the Prerequisite Viewer, click Close.

7. Repeat this procedure on all nodes in the cluster.

Step 3: Install management servers on the cluster nodes

Perform this procedure on every cluster node that will participate in the RMS cluster.

To prepare the cluster and install management server and user interface components

1. Log on to the cluster node that will be the primary owning node for the RMS with an account that is a member of the Administrators group.

2. On your installation media, start SetupOM.exe. This starts the System Center Operations Manager 2007 R2 Setup Wizard on the Start page.

3. Under the Install heading, click Install Operations Manager 2007 R2. This starts the Operations Manager 2007 R2 Setup Wizard.

4. On the Welcome page, click Next.

5. On the End User License Agreement page, select the I accept the terms in the license agreement option, and then click Next.

6. On the Product Registration page, enter the appropriate values in the User Name and Organization fields. Enter your 25-digit CD Key, and then click Next.

7. On the Custom Setup page, leave the management server and User Interfaces options set to This component, and all dependent components, will be installed on the local disk drive. Set the Database, Command Shell, and Web Console components to This component will not be available, accept the default installation location, and then click Next.

8. On the SQL Server Database Instance page, type the SQL Server name and database instance in the SQL Database Name box. This is in the format of SQL Server\SQL Instance. If the SQL Server database was installed in the default instance, you only have to enter the SQL Server 2008 name that is running the OperationsManager database.

9. Check that the SQL Database Name field reads OperationsManager.

10. Check that the SQL Server Port field has the value of 1433.

    Note

    If you have chosen a different port for SQL Server communications and have already configured that port in SQL Server, you should enter that value here; otherwise, accept the default value of 1433. If you have installed SQL Server using a named instance, type in the dynamic port value.

11. Click Next.

12. On the Management Server Action Account (MSAA) page, accept the default Domain or Local Computer Account option, enter the credentials of the MSAA, and then click Next.

    Note

    By using a domain-based account, it will be much easier to perform discovery and push agent installation later on than if you chose the Local System account.

13. On the SDK and Config Service Account page, select the Domain or Local Account option, enter the credentials for the System Center Data Access and System Center Management Configuration service account, and then click Next.

    Note

    In this configuration, the account must be a domain account, because reporting is installed on a separate server. This account must have permissions on the reporting system.

    Note

    If you receive an Account Verification Error when you click Next, it is most likely that you mistyped the credentials or the SDK and Config service account was not added to the local Administrators group.

14. On the Customer Experience Improvement Program page, optionally indicate whether you want to join this program, and then click Next.

15. On the Microsoft Update page, optionally indicate whether you want to use the Microsoft Update services to check for updates, and then click Next.

16. On the Ready to Install the Program page, click Install when you are ready for the installation to proceed.

17. On the Completing the System Center Operations Manager 2007 R2 Setup Wizard page, clear the Start the Console check box, and clear the Back up Encryption Key check box, and then click Finish.

    Important

    Even though the Operations console has been installed, do not launch the console at this point. Clear the Launch the Operations Console check box to prevent the Operations console from launching.

    Note

    If setup fails, it gives you a value to search on and a link to open the setup log.

18. Set the System Center Data Access service and the System Center Management Configuration service to start manually on all cluster nodes that will be in the RMS cluster.

Step 4: Prepare the RMS cluster resources

In this procedure, you create cluster resources out of the System Center Management service (HealthService), the System Center Management Configuration service (OMCFG), and the System Center Data Access service (OMSDK). These are the RMS resources that can fail over between cluster nodes along with the network name, IP address, and physical disk.

To prepare the RMS cluster resources

1. Log on to the node that is the owner of the RMS application with an account that is a member of the Administrators group.

2. In Failover Cluster Management, in the navigation pane, right-click the RMS application, and then select Add a resource 4-Generic Service.

3. On the Select Service page, select the System Center Management Service, and then click Next.

4. On the Confirmation page, click Next.

5. On the Summary page, click Finish.

6. In the summary pane of your RMS application, right-click the System Center Management resource, and open its properties.

7. On the Dependencies tab, click Insert, and select the shared disk that you prepared for the RMS cluster from the Resource list.

8. Click Insert again and select the network name from the Resource list.

9. Click Apply.

10. Click OK.

11. Repeat the same process for the System Center Management Configuration service and the System Center Data Access service.

Warning

Do not bring these services online at this time.

Step 5: Create the virtual RMS

To create the virtual root management server

1. Ensure that the file share with the encryption key is accessible by all cluster nodes. This is used for distributing the RMS Key.

2. If you successfully backed up the encryption key at the end of RMS setup, you can skip to step 5.

3. On the stand-alone RMS , as an Administrator, open a Command Prompt window and change the current directory to the Operations Manager installation folder; for example, cd\Program Files\System Center Operations Manager 2007.

4. To back up the RMS Key, type the following, where is a share accessible to all cluster nodes:

   SecureStorageBackup.exe Backup \\\.bin.

   Note

   This starts the Encryption Key Backup or Restore Wizard. Enter a password. It must be at least eight characters long and must include at least one symbol. You must confirm the password to create the encryption key file.

5. Log on to each secondary management server computer that is a node in the RMS cluster with an account that is a member of the Administrators group.

6. On each secondary management server, navigate to the Operations Manager installation directory and start SecureStorageBackup.exe as an administrator.

7. This starts the Encryption Key Backup or Restore Wizard. On the Introduction page, click Next.

8. On the Select Action page, select the Restore the Encryption Key option, and then click Next.

9. On the Provide a Location page, type in the path or browse to the encryption key file, and then click Next.

10. On the Provide a Password page, type in the password that you used when you backed up the encryption key, and then click Next.

11. Click Finish.

12. Repeat this procedure on each cluster node that will be participating in the RMS cluster.

13. On the computer running SQL Server and hosts the OperationsManager database, open the SQL Server Management Studio tool, expand the Databases folder, and select the OperationsManager database. Right-click, and then select Tasks, and then click Back Up to start a backup. On the Back Up Database - OperationsManager page, ensure that the Backup type value is set to Full, give the Backup set an appropriate name, and set the Backup set will expire value to a date in the distant future. In the Destination box, for the Back up to value, select Disk and add an appropriate disk location to hold the backup, if one is not already present, and then click OK.

    Important

    When you run the ManagementServerConfigTool tool to create the RMS cluster, you are advised to back up the OperationsManager database because irrecoverable damage can be done by creating the RMS cluster if something is done incorrectly.

14. Log on to the RMS application-owning node with an account that is a member of the Administrators group and is in the Operations Manager Administrator role.

15. In Failover Cluster Management, expand the cluster, and ensure that the RMS application is owned by the node that you are logged on to.

16. Navigate to the Support Tools directory on your source media and locate the operating system architecture folder that corresponds to the operating system architecture that is on the cluster nodes (AMD64 or i386 or ia64).

17. Copy the ManagementServerConfigTool.exe file to the System Center Operations Manager 2007 installation directory.

18. On the pre-existing RMS, open the Operations console and navigate to the Administration view.

19. In the Administration view, open the Management servers container which is under Device Management. In the results pane, you should see the pre-existing RMS with the value of Yes in the Root Management Server column. You should also see all of the cluster nodes that you installed secondary management servers on listed and with the value of No in the Root Management Server column.

    Important

    Check that the System Center Data Access and the System Center Management Configuration services on all cluster nodes have their startup type configured as Manual.

20. On the RMS application owning node, as an Administrator, open a Command Prompt window, type cd , and then press ENTER.

21. To instantiate the RMS cluster group as a cluster, type the following, where G is the disk resource that is allocated to the cluster group that is being used to create this virtual root management server and where is the network name resource allocated to the same cluster group:

    ManagementServerConfigTool.exe InstallCluster /vs: /Disk:G

    The value you enter for must be the value that appears in the Name box on the General tab of the Properties dialog box.

22. You will be prompted with this message: Running this tool can cause irreversible damage to your Operations Manager DB. Please backup your DB before continuing. Continue the Install Cluster action? (Y/N):. Type Y.

    Note

    ManagementServerConfigTool.exe InstallCluster installs the RMS as a clustered service on every available node in the cluster and demotes the pre-existing RMS to a management server.

23. In Failover Cluster Management, right-click the RMS application and click Bring this Service or application Online to bring all the RMS applications online.

24. Open Failover Cluster Management and right-click the RMS application. Select Move this service or application to another node and select the next node in the cluster. Repeat this so that the RMS application is moved to each node of the cluster.

    Important

    The RMS application must be moved and come online successfully on each cluster node to set the state of the services correctly on each node at this time. Do not skip this step.

25. Restart the System Center Management Service on the pre-existing RMS. This is necessary so that it recognizes that it has been demoted from RMS to a management server.

26. Open the Operations console. You will be prompted to connect to the RMS. Type in the name of the clustered RMS.

27. Open the Administration view and expand the Management Servers container under Device Management.

28. In the results pane you should see the server that was the pre-existing RMS listed and with a value of No in the Root Management Server column. This indicates that it has been successfully demoted to a secondary management server. You should see the VirtualServerNetBIOS name of the clustered RMS listed and with a value of Yes in the Root Management Server column. You should not see the individual cluster nodes listed at all.

    Warning

    If you do not see these results in the Operations console, the InstallCluster action did not complete successfully, and you have to follow the steps in the “Recovering from a Failed InstallCluster Action” procedure.

The clustered RMS setup is completed.

To recover from a failed InstallCluster action

1. Shut down all the cluster nodes that are hosting management servers.

2. Shut down the pre-existing RMS server.

3. Logon to the SQL Server that hosts the OperationsManager database with SQL Administrator rights and follow the How to Restore Operations Manager 2007 Databases (http://go.microsoft.com/fwlink/?LinkId=180782) procedure to restore the OperationsManager database.

4. After the OperationsManager database has been successfully restored, start the pre-existing RMS, and open the Operations console.

5. Examine the Alerts view for any alerts regarding the pre-existing RMS, and fix the issues indicated in the alerts, and then resolve the alerts.

6. Wait 15 minutes to see if any new alerts are raised regarding the pre-existing RMS. If no alerts are raised, and the pre-existing RMS show a green health state, proceed to the next step; otherwise, repeat steps 5 and 6 until there are no alerts regarding the pre-existing RMS and its health state is green.

7. In the Operations console, navigate to the Management Servers container and delete all the cluster nodes that are running secondary management servers.

8. Start all of the cluster nodes that are secondary management servers and log on with an account that is a member of the Administrators group.

9. Uninstall all Operations Manager 2007 R2 components via Control Panel, Programs and Features.

10. On one of the cluster nodes, open the Failover Cluster Management tool and select the RMS application that you previously created. Ensure that the application is owned by the node that you are on.

11. In the results pane, under Other Resources, delete the System Center Data Access, the System Center Management and the System Center Management Configuration resources.

12. Resume and follow all steps starting with the Installing Management Servers on the Cluster Nodes procedure earlier in this topic.

Step 6: Register the SPN

Use this procedure to correctly register the SPN of the RMS.

To use SETSPN

1. Log on to the owning node with an account that is a member of the Administrators group in the domain.

2. Open a Command Prompt window using the Run as administrator option.

3. Use the following command, where \ is the account that the Failover Cluster service is running under, where the value for is the same name you use for with the domain extension appended, and where the value for must be the value that appears in the Name box on the Parameters tab of the Properties dialog box for the network name resource: runas /user:\ "SetSPN.exe -A MSomHSvc/ "You must do this for both the FQDN of the virtual RMS server and for the NetBIOS name. For example: SetSPN –A MSomHSvc/cRMS cRMS will set the NetBIOS name and SetSPN –A MSomHSvc/cRMS.lab.demo cRMS will set the FQDN.

4. In the Command Prompt window, type setspn –L . In the list that appears, if the MSomHSvc SPN has been registered correctly, you see it listed. For example, you should see: MSOMHSvc/MSOMHSvc/

5. If the SPN is listed, but is incorrect, run SetSPN -D MSOMHSvc/ to delete it, and then run the SetSPN -A MSOMHSvc/cRMS cRMS and SetSPN -A MSOMHSvc/cRMS.lab.demo cRMScommands to register it.

Step 7: Prepare an inaccessible management server

The following procedure is required only if, in bringing the cluster group online, the output stated that you are required to run the ManagementServerConfigTool.exe tool by using the AddRMSNode action on any of the non-root management server cluster nodes. This is most likely caused by a cluster node that is not accessible when the InstallCluster action was executed, or you are adding a new node to the cluster.

To prepare inaccessible management servers

1. Log on to the computer that hosts the management server as a member of the Administrators group.

2. Open the services snap-in and if the startup type for the System Center Data Access Service is set to Disabled, change it to Manual.

3. As an administrator, open a Command Prompt window, change directories to the installation folder, and type the following:

   ManagementServerConfigTool.exe AddRMSNode /vs: /Disk:

   VirtualServerNetbiosName is the Network Name resource allocated to the same cluster group. The value you enter for VirtualServerNetbiosName must be the value that appears in the Name box on the General tab of the Properties dialog box for the Network Name Cluster resource.

   VirtualServerDiskResource is the disk resource allocated to the cluster group being used to create this virtual root management server. The Disk location can be found in the results pane of the properties for the RMS application.

Step 8: Test the cluster installation

Use the following procedure to test the cluster installation.

To test an Operations Manager 2007 cluster

1. In the Operations console, click Administration.

   Note

   When you run the Operations console on a computer that is not a management server, the Connect To Server dialog box appears. In the Server name text box, type the name of the RMS Server (the cluster virtual server name) that you want the Operations console to connect to.

2. In the Administration pane, expand Administration, expand Device Management, and then click Management Server.

3. In the Management Servers pane, the RMS Server Network Name should appear with a health state of Healthy.

4. In the Administration pane, click Agentless Managed.

5. In the Agentless Managed pane, the entry for each node in the cluster should appear with a health state of Healthy.

Post-Procedure Checklist

After you successfully complete the RMS clustering procedure, perform the tasks in the following checklist.

• Re-enable notifications.
• All agents that were reporting to the pre-existing RMS have been redirected to the clustered RMS. If this is not the configuration that you want, you must manually reassign the desired agents to another management server in the management group.
• If you have System Center Operations Manager 2007 R2 Reporting installed, you must configure it to point to the new RMS.
• If you have System Center Operations Manager 2007 R2 Reporting installed, you must configure the OperationsManagerDW database to point to the new RMS.
• If you have the System Center Operations Manager 2007 R2 Web console installed, you must configure it to point to the new RMS.

To configure the reporting server with the name of the new root management server

1. Log on to the reporting server.

2. Navigate to the installation folder of SQL Server Reporting Services, for example, %ProgramFiles%\Microsoft SQL Server\MSSQL.2\Reporting Services\ReportServer if you are using SQL Server 2005 or %ProgramFiles%\Microsoft SQL Server\MSRS10.MSSQLSERVER\Reporting Services\ReportServer, if you are using SQL Server 2008.

3. Open the rsreportserver.config file in Notepad, and locate the two instances of ServerName, where ServerName is the name of the original root management server. Change ServerName to be the name of the new root management server.

4. Save the file, and then close Notepad.

5. Open the registry and locate the key HKEY_LOCAL_MACHINE\Software\Microsoft\Microsoft Operations Manager\3.0\Reporting.

6. Change the DefaultSDKServiceMachine value to be the name of the new root management server.

To configure the data warehouse server with permissions for the new root management server

1. On the server hosting the data warehouse, open Microsoft SQL Server Management Studio, and in the Object Explorer pane, navigate to Security, and then expand Logins.

2. Locate the account that corresponds to the new root management server and on which the System Center Data Access Service is running (if it is running under the Local System account, the format is <domain\computername$>).

3. Right-click the account and select Properties.

4. In the Login Properties dialog box, in the Select a page pane, select User Mapping.

5. In the Users mapped to this login list, in the Map column, select the box that corresponds to the OperationsManagerDW database.

6. In the Database role membership for: OperationsManagerDW list, ensure that the following items are selected: configsvc_users, db_datareader, db_datawriter, db_ddladmin, and sdk_users.

7. Click OK to save your changes and to close the Login Properties dialog box.

To configure the Web console with the name of the new root management server

1. Log on to the Web console server.

2. Navigate to the installation folder of the Web console (by default, %ProgramFiles%\System Center Operations Manager 2007\Web Console).

3. Open the Web.config file in Notepad.

4. Locate the line ‘’, where RootManagementServer is the name of the original root management server. Change RootManagementServer to be the name of the new root management server.

5. Save your changes, and then close Notepad.

Original Post can be found here

So, if you are in the position of needing to install SIS on a Windows Server 2008 system to meet the prerequisites of DPM, run the following command at an Administrator Command Prompt:

start /wait ocsetup.exe SIS-Limited /quiet /norestart

The beauty of this is that it will return you to the Command Prompt within seconds and provide you with no indication that SIS was installed successfully. Job well done on this piece of software! :)

In part 1 of this series, Virtual Operations Manager: Part 1 - Should I Virtualize?, we discussed which Operations Manager 2007 roles could and should be virtualized. From this point, we're going to broaden the discussion to virtual workloads as a whole, with info applicable to all System Center products. Really the discussion we're about to have here is just the tip of the storage iceberg. To get the complete storage picture, we have to look at storage drivers, storage controllers and the underlying options for storage architecture. But we'll leave some of these deep discussions of storage for another day.

In This Article

In this article, I will provide some authoritative source material we found in researching this issue in the past few months. I'll provide recommendations and source material for

1) The Case for Thin Provisioning

2) Recommendations from Microsoft (Hyper-V) and VMware (ESX / vSphere) for production workloads

3) Quantitative Information on the difference in performance of dynamic versus fixed virtual disks.

4) Simple, real-world, common sense guidance to help you in choosing which virtualized production workloads (if any) should be thin provisioned.

The Case for Thin Provisioning

First, let's just address why would you use dynamic VHDs in the first place. If I know fixed size disks are faster, why do I want to thin provision? The answer is simple - to save storage space. If you don't care about utilization of your storage, then you don't need to use dynamic disks. But let's face it, SAN storage is not free, nor are the storage architects and engineers we have to pay to design, deploy, migrate, update and tend to this infrastructure. We all want to save storage space.

In Hyper-V, you can create a dynamic VHDs. In VMware thin provisioning can be leveraged within the storage array as well as at the virtual disk level. We're going to focus on the virtual hard disk as common element here (VHD in Microsoft-speak, VMDK in VMware terms).

Microsoft (Hyper-V)

What is Microsoft's recommendation for use of dynamic VHDs in production workloads? We can look at a couple of good sources within MS for answers. There is a short blog post that addresses this question, titled Hyper-V  Dynamic or Fixed Size VHD , by Mark Ghazai [MSFT].

Mark says this:

• Best practice says; “Dynamic expanding VHDs are not recommended for virtual machines that run server workloads in a production environment”.
• Dynamic expanding VHDs will suffer a slight performance impact on writes when a dynamic VHD has to expand. This is one of the reasons why it doesn’t give the exact same performance as a fixed size VHD.

However, both disk types are now supported by Microsoft in production. However, the only reason someone would use Dynamic disks is if they’re trying to save space otherwise I’d still recommend using Fixed VHDs. You can always expand the fixed size VHD (Using Edit Disk option in Hyper-V management Console or PowerShell script) if you need to.

Performance Comparison

The real question is "how much better is slightly better"? What do I lose with dynamic VHDs?

Tim Litton [MSFT] answers this question in Hyper-V and VHD Performance - Dynamic vs. Fixed.

Fixed VHD always performs better than a Dynamic VHD in most scenarios by roughly 10% to 15% with the exception of 4k writes, where Fixed VHD performs significantly better, as shown in this chart.

How do you know where your workload falls in this chart?

Do your own testing. As Tim says "The exact result that a customer is going to see will depend on quite a few variables (e.g. how large and often the reads and writes are, how many outstanding I/O there can be at one time), so performing real-world testing is the best way to assess what impact virtualization will have."

VMware (vSphere) 

VMware has plenty to say on this topic as well. VMware has a whitepaper titled "Dynamic Storage Provisioning, Considerations and Best Practices for Using Virtual Disk Thin Provisioning"

Excerpts: When considering use of thin provisioned virtual disks there are often concerns about its impact on overall virtual machine performance. These concerns often stem from the perception that fragmentation may reduce performance when the entire disk is scanned as well as more frequent updates to the metadata updates may cause increased contention. The next portion will address why both those concerns should not prevent one from considering use of thin virtual disks.

Thin provisioning won’t be an ideal solution for every environment. But when used and managed correctly, it does offer significant benefits to providing efficient and dynamic space allocation that improves the overall operations for many virtual administrators. 

Common Sense Guidance

As you can see, the answer is no longer simply "no". However, I recommend that you only use thin provisioning in specific scenarios.  

• Don't use thin provisioning if you're not concerned about storage capacity...(but who isn't pushing storage to the limit these days?)
• Do use thin provisioning for data (non-database) and application binary volumes. Particularly, server roles that are light on disk I/O will be good candidates (like the OpsMgr Gateway role).
• Don’t use thin provisioning for the system volume of a VM, database-type data partitions, or volumes that will grow data rapidly.

For specific guidance on System Center workloads (OpsMgr 2007 specifically), have a look at Virtual Operations Manager: Part 1 - Should I Virtualize?

Conclusion

I hope you've found some value in this post. I have another post lined up on the deeper discussion of storage architecture. I'll try to follow up with "Virtual System Center Part 3: Storage Architecture" as soon as I can. With MMS 2010 only a few weeks ago, spare time is in short supply, so may be a bit. If you're headed to MMS 2010, I'll see you in Vegas!

These short videos focus on specific tasks and show you how to accomplish them in System Center. Subscribe to the System Center videos RSS feed to see new content every month. Click here for the link

Nice! ]]> Thu, 11 Mar 2010 02:05:44 GMT /BlogDetails/tabid/143/IndexId/60713/Default.aspx Tenchuu has an interesting view on life, I am surprised that he he only a ‘Junior Application Engineer’ some would say that he will be in time a great SDK Chap. (views of SCC)

First Name: Andreas

Last Name: Zuckerhut

Forum User Name: Tenchuu

Which country do you live in?

Austria, Vienna

What is your primary role at your work?

My title is Junior Application Engineer but my primary role is: Management Pack Developer.

What is you greatest and proudest IT achievement?

The Management Pack automation using the SDK I created here: Create MPs upon specific server types including groups with dynamic membership, views and also automatic creation of Event rules and a new Flooding prevention for them. The old one is a joke compared to it.

Outside of computers what are your interests?

My girlfriend, computer games, Poker, lol stuff, movies, TV series... entertainment I’d say. Sometimes I’m into crafting, or MSPainting something that people can laugh about.

I’m currently building a Super Mecha Death Christ Action Figure including Sound Buttons so it can swear and say: “Watch the ******* language!”

It’s not like I’m really good at it, but I like to build things every now and then. It’s a nice compensation when you are used to building castles in the air. With that I mean: IT stuff. You can’t touch the things you set up or program in your daily business.

If you were stranded on a desert island what would you most want to have with you?

John Lock. I want to survive of course and he’s the man.

What’s the craziest thing you have done in your life to date (that we can publish)?

I was sliding with approximately 50 km/h over asphalt on my socks. Not that I wanted to do that, but when I sat down on a friend’s rusty car’s bonnet, she didn’t like it and started accelerating. And then she jumped on the break. Yea, she’s a total nut but I guess the poisonous vapours are to blame. She’s an automotive painter.

Nothing bad happened anyway so I’m fine with it.

If you were a Star Trek of Star Wars character, which one would you be?

Khan maybe, but only because I sometimes use the name: Tenchuu Khan. I have no clue what kind of guy Khan is ;)

If you could trade places with any other person for a week, famous or not famous, living or dead, real or fictional. With whom would it be?

I’d not trade places with anyone. When I was a teenager, I guess I would have named some game/comic hero. It’s not like I think I’m perfect, it’s more that everyone has to pay the prize for being who he is. So, I really don’t get the point of trading places with anyone. I’m fine with where I’m at currently.

Who is your favourite singer or group?

That’s a tough one. To be honest, when I like a song or a band I listen to it all day until I can’t hear it anymore. But a song I listen to at least once per month: Donovan – Hurdy Gurdy Man.Currently I’m running the Amelie piano theme and Carly Comando – Every Day in a loop ;)

What’s your favourite film?

Same problem here. There are way too many. However, I think the best movie ever is: Forrest Gump

If aliens landed in front of you and, in exchange for anything you desire, offered you any position on their planet, what would you want?

If that would happen, I guess they should have a plan for me anyway. Why else would they travel here, picking me up? I’m not into power and leadership anyway... unless I’m playing Total War.

How would you react if you are transformed into a fish?

I guess I’d swim around?

If you could meet anyone (dead or alive) who would it be and why?

James Rolfe. I’m a cinemassacre.com fan and he seems to be a guy you can drink a beer with and talk about games/movies.

Thank you for your time and effort you have put into System Center Central, we all really appreciate it. SCC