This solution has the ability to create any Visio diagram and link any item on that diagram to object of a class to show it’s health. I don’t use SharePoint integration, but probably someday I would, rather that I use only full screen view (F5 key) in Visio to maximize my map. As I’m not a great graphic, my diagrams are rather straightforward.

Now let’s get into the main point. As a title of my blog says, there are some problems. 

1. For the requirements there are 3 of them, where one in a requirement for second one. So you need .NET Framework 3.5.1 (which is actually enabled in Windows 7 as a feature), OpsMgr 2007 R2 Console or Authoring Console and a Visio 2010 Premium or Professional. When you install everything and try to install Visio Add-In, you might get a pop-up. that there’s no Visio installed… Well… how come if I installed it just before? The point is – you have to run Visio for the first time and close it – Visio Add-In now can be installed.

2. After installation you and after linking or adding an object, you don’t see the health indicators (icons, color etc.). This is reflected to be a bug in add-in probably because of differences in language versions of OS and Visio. If so – go to C:\Program Files (x86)\Visio 2010 Add-in for System Center Operations Manager 2007 R2 and run masters.vhd – that enables colorful icons and fillings.

Thanks for Cleber Marques for helping me with this (though I found it myself ).

The process started at 07:18:53 failed to create System.PropertyBagData. Errors found in output:

C:\Program Files\System Center Operations Manager 2007\Health Service State\Monitoring Host Temporary Files 1\9965\SCOMpercentageCPUTimeCounter.vbs(125, 5) SWbemRefresher: Invalid class

Command executed: “C:\Windows\system32\cscript.exe” /nologo ”SCOMpercentageCPUTimeCounter.vbs” server.domain.name false 3

Working Directory: C:\Program Files\System Center Operations Manager 2007\Health Service State\Monitoring Host Temporary Files 1\9965\

One or more workflows were affected by this.

Workflow name: Microsoft.SystemCenter.HealthService.SCOMpercentageCPUTimeMonitor

Instance name: server.domain.name

Instance ID: {7CA9600A-745C-DDF6-0211-88A22E13C775}

Management group: MGMT_GROUP

Running WBEMTEST, connecting to root/cimv2 and enumerating your classes recursively will give you about 800 classes instead of more than 1000. This is because Win32_PerfFormattedData and Win32_PerfRawData are not registered. To re-register it, you’ll have to rebuild your performance counter dll’s from scratch. You can do this by creating a bat or cmd script with such code:

@echo off
sc config winmgmt start= disabled
net stop winmgmt /y
%systemdrive%
cd %windir%\system32\wbem
for /f %%s in (‘dir /b *.dll’) do regsvr32 /s %%s
regsvr32 %windir%\system32\tscfgwmi.dll
wmiprvse /regserver
winmgmt /resyncperf
net start winmgmt
net start iphlpsvc
net start CcmExec

Then run this on server and you’ll have then no worries. You don’t have to reboot server nor restart agent – everything will start to work immediately. Basically last two lines mean that services are basing on Windows Management Instrumentation service. So you have to start them too as they won’t restart. I have here IP Helper and SMS Agent Host. So if you have more of them, just add them at the end of the script.

And if you want to read more about problems with WMI, check this fantastic post: http://www.paessler.com/knowledgebase/en/topic/3713-i-have-tried-a-lot-of-things-to-fix-my-wmi-what-else-can-i-try-to-avoid-reinstalling-windows

http://blogs.technet.com/b/momteam/archive/2011/10/11/opsmgr-2007-r2-now-supports-sql-2008-r2-sp1.aspx

Is it worth?

Yes, although you have to be sure that Marnix Wolf, Kevin Holman or other popular MVPs will post bugs found in that one (as it happened in OS MP – details here). Let’s check what’s new inside. Below is an outtake from official AD MP Documentation:

• Fixes to problems reported by customers:
• Active Directory databases larger than 4 GB reported incorrectly.
• 20% of the alerts are not triggered due to wrong event ID mapping.
• Performance data is not collected due to wrong event ID mapping.
• Performance counter selected by default is wrong.
• Time skew alert is not triggered due to script defect.
• Operation master monitor is broken due to script defect.
• Frequent operation master alert description misspelled.
• Fixes to architectural issues to facilitate future System Center Operations Manager releases:
• Discovery interval for client perspectives set to larger values.
• Discovery scheduler class is used on several discoveries.
• Views target a custom AD DS MP class instead of System.Entity.
• Reports target a custom AD DS MP class instead of System.Entity.
• Some discovery targets will not change Properties.

So as the console was still open, I’ve decided to check, how is it going. You can import it via Management Pack Catalog from Import Management Pack wizard.

As you see… [ekhm]… misspellings where corrected

There are more changes to scripts, monitor roll-ups and there are some undocumented overrides for Read-Only Domain Controllers, so please check the whole documentation. After importing the MP – nothing happened. My console is still alive, alerts are ok, performance is ok. Going to look for mistakes now

Download the NEW AD Management Pack!

1: Uruchomienie raportów może się nie udać z powodu błędnych ustawień bezpieczeństwa procedur składowanych;

2: Import MP może się nie udać, jeśli w poprzednim MP było wiele override’ów na miejsce na dysku logicznym; błąd spowodowany… literówką;

3: Baza wiedzy nie jest wszędzie uaktualniona do nowych wartości domyślnych w Management Packu;

4: Monitorowanie BPA powoduje często zalew alarmów (kto trzyma AD w zgodzie z BPA ręka do góry!);

5: Raport “Performance by utilization” po zaznaczeniu licznika Logical Disk % Idle time jest do góry nogami: najniższe wartości są na górze (100% Idle time jest najniższa) a najwyższe opóźnienia na dole (wszystko około 10% lub mniej).

Cały temat znajduje się na witrynie Connect: https://connect.microsoft.com/OpsMgr/feedback/details/692944/new-server-os-mp-version-6-0-6957-0-has-some-bad-bugs – jest już 50 potwierdzeń błędu i to jest chyba największa tego typu zorganizowana akcja przez społeczność OpsMgr.

• Optimized operating system performance collection rules.
• Added two new operating system-specific reports:
• Performance by System
• Performance by Utilization
• Added support for the Server Manager Best Practices Analyzer (BPA). The monitoring pack now collects BPA results from monitored servers and returns the BPA state to Operations Manager.
• Added support to discover and monitor Cluster Shared Volumes (CSV).
• Modified collection rules to include the latest service packs and hotfixes.
• Changed the threshold for the “Average Disk Seconds Per Read”, “Average Disk Seconds Per Transfer”, and “Average Disk Seconds Per Write” monitors to 0.04.
• Added new monitors for “Disk $ Idle Time” and “Current Disk Queue Length”.
• Changed the monitor types for disk performance from Average Samples to Consecutive Samples to reduce noise and improve collected data.
• “Logical Disk Availability Monitor” has been renamed to “File System Error or Corruption”.

Rzuca się w oczy dodanie monitoringu BPA, Cluster Shared Volumes i usprawnione monitorowanie dysków fizycznych. Nie pozostaje Wam nic innego jak uaktualnić MP do najnowszej wersji z poziomu konsoli Administration i importu MP z katalogu (wybierając filtr – Updates to imported Management Packs) lub też poprzez pobranie i instalację MP na dysku i wgranie standardowe. Została też uaktualniona dokumentacja, która zawiera bardzo szczegółowy opis wszystkich monitorów, reguł i grup, które są zawarte w Management Packu.

http://www.microsoft.com/download/en/details.aspx?id=9296

EN: This is the list we’ve been waiting for. Now, when I have more time and ability, it’s time to do some refreshing. Below is a list of network devices that are fully supported by Operations Manager 2012 Network Monitoring module.

Widzimy się 19 września w siedzibie Microsoft Polska w Warszawie z samego rana i jedziemy do 16:30!