I saw a question about generating a report every hour for “the hour before”, that you can´t do out of the box. But I have a idea, why dont generate a report every half hour, for the entire day? That would give you a updated report every half hour, including the hour before, or as far as the data warehouse has processed data.

In this example I will use a generic performance report

1. Open the Ops Mgr 2007 Console
2. Go to Reporting (Ctrl+3)
3. Expand Reporting and click Microsoft Generic Report Library
4. Click Performance and then choose Schedule from the action menu
5. The Subscribe to a Report wizard starts, at the Delivery page, input a description and choose a delivery method. In this example I will choose Windows File Share as delivery method. That will store my report to a file share on the network.
   – File name, for example CPU Graph Today.pdf
   – Path, for example \\servername\share
   – Render Format, choose output format, PDF is nice
   – Write mode, choose Overwrite
   – User name and Password, inpute information about a account to use to write the report, then click Next
6. At the Subscription Schedule page, choose how often the report should be generated. I will choose Hourly and every 0 hours and 30 minutes. Choose when the subscription should start and then click Next
7. At the Report Parameters page, choose
   – Data Aggregation: Hourly
   – From: Today 00:00
   – To: 23:59
   – Histogram: Daily by hours
   – 3D Chart: True
8. At the Report Parameters page, click Change…
9. In the Settings box, click New Chart
10. In the Settings box. input chart title, for example CPU Graph Today
11. In the Settings box, click New Series
12. In the Settings box, click Add Objects, in the Add Object box, choose a item of the Windows Operating System type, click Add and then OK
13. In the Settings box, click Browse
14. In the Select Rule box, search for “processor” and choose “Processor % Processor Time Total” from available items, click OK
15. In the Settings box, click OK
16. At the Report Parameters page, click Finish
17. Back in the console reporting part, click Scheduled Reports to see your new subscription.

As you can see in the report, the report are about one hour behind. If you generate the report at 23:30 the report will include data to 22:30. If you need for example performance data quicker than that, you can create a new performance view.

In this post I will show how to create a monitor that check for one event and if there is not another event within a specified timeframe minute from the first event, the monitor will generate an alert. I will reset the monitor with time, 3 minutes, but you can choose to reset the monitor with for example a third log event.

1. Start the console
2. Go to Authoring, expand management pack objects and click Monitors
3. Click Scope and select Windows Computer, click OK
4. Expand Windows Computers, expand Entity Health, right-click Availability and choose to create a new unit monitor
5. Create a unit monitor – Monitor Type: Choose Windows Events/Correlated Missing Event Detection/Timer Reset, click Next
6. Create a unit monitor – General: Input a name and a description, click Next
7. Create a unit monitor – Missing Event Log Name A: Input the event log name of the first event, click Next
8. Create a unit monitor – Build Missing Event Log Expression for A: Input event ID and event source, in my example it will be event id 1000 and event source EventCreate. Click Next
9. Create a unit monitor - Missing Event Log Name B: Input the event log name of the second event, click Next
10. Create a unit monitor – Build Missing Event Log Expression for B: Input event ID and event source, in my example it will be event id 2000 and event source EventCreate. Click Next
11. Create a unit monitor – Configure Correlation:
Correlation interval: 1 Minutes
Correlation Details: The last occurrence of A with the configured occurrence of B in chronological order
Click Next
12. Create a unit monitor – Auto Reset Timer: In my example I will specify 3 minutes, click Next
13. Create a unit monitor – Configure Health: Click Next
14. Create a unit monitor – Configure Alerts: Check “Generate alerts for this monitor” and then click Create

If I only get a event ID 2000 and no event ID 1000 there will be a alert. If I get event ID 2000 and event ID 1000 within 1 minute there will be no alert. You can change the correlation configuration in any way you want, for example in which order the events must be generated.

Ops Mgr product team at Microsoft has made a tool to work with proxy settings in Ops Mgr 2007. With this tool you can easy browse proxy settings. A great way to for example very that all your domain controllers has the correct setting. Current version is 1.1.

You can download the tool here.

Boris Yanushpolsky in the Ops Mgr product team at Microsoft has made a tool to work with overrides in Ops Mgr 2007. With this tool you can easy browse overrides based on type or computer.

You can download the tool here , current version is 3.2

I saw a question today about monitoring scheduled tasks with Operations Manager 2007. If you open the scheduled task window (start menu > all programs > accessories > system tools > scheduled tasks) you can open a logfile for your scheduled jobs, from the Advanced drop down menu, the default file is C:\Windows\SchedLgU. You can monitor this file from Ops Mgr 2007, and search for words like WARNING and ERROR. In this post you can read how to setup monitoring of a logfile like that.

I want to inform you about some great information that was recently uploaded to Systemcenterforum.org.

This download contains two Excel spreadsheets containing the shared script mappings for the Active Directory and Exchange management packs, including script name, monitors and rules using the script, as well as object type targeted. Tables list only scripts used by more than one workflow.

Download the file here