OpsMgr and empty reports – Part 4 – Examples for disk reports

----------------------------------------------------------------------------------
Postings in the same series:
Part I   – The Introduction.
Part II  – Tips & Tricks
Part III – Targeted Reports 
Part V  – Scheduling / Publishing Reports and some tricks
----------------------------------------------------------------------------------

On the previous postings of this series I got a lot of mail and comments. Many of which were about reports for disks like Physical Disk Performance History, Performance Reports on Logical disks and so on. Somehow many people don’t get filled reports for these items. So this posting will be about those reports and how to get them filled.

Before I start I must say that this posting is based on information from Cameron Fullers article about making a free disk space report. So when you’re new to that article read it and then come back since this posting will be – as much as possible – in telegram style.

1. Go to Reporting > Microsoft Generic Report Library > double click on the Performance report. The Performance Report will be opened.
   
   
2. In the parameter area click on the button Change. The settings screen will be opened now. This is the most important screen for this report since here you can select anything you want for a Performance Report.
   
   
   
3. Click on New Chart and then on New Series. This way a new chart will be added to the List area with one item:
   
   
   
4. Click the Add Group button. And now here comes the most important part of all: the selection of the component. As stated in in the second posting of this series, under the header ‘Add Group vs. Add Object’: ‘…When running a Report it is better to use the ‘Add Group’ option and select the server(s) with the Object Class Windows Computer’. This way the report will most of the times be filled because Windows Computer has a parent relationship to almost all other classes, also the class needed to get a filled report.’
   
   And here the same rule applies. So select the Windows computer, in this case dc01.systemcenter.org:
   
   Be sure to select the Class Windows Computer. Click Add so it appears in the Selected Objects pane and click OK.
   
   
5. Only a counter has to be selected in order to get a filled report. For this, click on the Browse button nearby the Rule section:
   
   
   
6. Go to the second tab, Search by Counter and select as Performance Object PhysicalDisk.  Leave the Counter on (All) and click on Search. All available counters will be displayed:
   
   
   
7. For this example I have chosen the first one, Physical Disk Average Disk Queue Length 2008. Click OK. The screen as shown in step 3 is to be seen now. Select Specific Instances and select the first drive (C:).
   
   
   
   
8. Then click New Series button and repeat steps 4 to 7 but now select the D: drive (as long as it is a separate physical drive. If it is not the case then select at step 7 All Instances). This screen will be shown when completed:
    
   Repeat steps 4 to 7 for every physical disk you want this information for. Click OK when done.
   
   
9. Select the timeframe, whether or not to use business hours and the type of graph.
   
   Adjust it as needed and click Run.
   
   
10. The report will be like this: 
     
    

This example can be used for any other reports as well. It goes beyond this posting (and my blog as a whole) to describe every report which can be created in the same manner. Just experiment and have fun with it. When you get the hang of it, you’ll see there is a lot more in the Data Warehouse then one expects. And when it is in the DW, it can be put into a report.

Also, when a report runs as it should, give the chart in step 3 a nice name since it will be shown in the report. Something like Physical Disk Average Disk Queue Length Drive 0 Server DC01. Or just for fun and the IT Manager without any real knowledge: 

Of course, a single report can contain more graphs then just one. For instance, a report showing the disk, cpu, nic and RAM performance of a single server. And when the report is doing just fine, Publish it. Give the report a good name and a description. For the start date is better to use offsets like 7 days or 2 weeks for instance. In an upcoming blog posting in the same series I will talk about scheduling reports and share some tricks how to go about it.