Friday, July 17, 2015

Dell OpsMgr MP Suite Unit Monitors

If you have installed the Dell SCOM MP Suite and are wondering why some of the columns may not be lighting up, this post may help.

Here is the link to the Dell Server Management Pack Suite v6.1 for System Center Operations Manager User's Guide because I will reference it several times in this post.  http://topics-cdn.dell.com/pdf/dell-srvr-mgmt-pck-v6.1-mscom_User's%20Guide_en-us.pdf

First I will give you some page numbers to see charts for the different monitoring features so you can see and make decisions on what you want to monitor and view.

Agent-based:
Comparison of Scalable and Detailed on page 10.
Discoveries by the Agent-based Monitoring Feature (Scalable & Detailed) starting on page 11.
Monitored hardware components (Scalable vs. Detailed) starting on page 12.
Unit Monitors (Scalable vs. Detailed) starting the bottom of page 19.

Agent-free:
Comparison of Scalable and Detailed on page 33.
Discoveries by the Agent-free Monitoring Feature  page 35.
Monitored hardware components (Scalable vs. Detailed) starting on page 36.
Unit Monitors (Scalable vs. Detailed) starting the bottom of page 41.

DRAC:
Discoveries by the DRAC Monitoring Feature on page 50.
Unit Monitors starting the bottom of page 52.

Chassis Monitoring Feature:
Discoveries by the Dell Chassis Monitoring Feature on page 58.
Monitored hardware components (Scalable vs. Detailed) starting on page 58.
Unit Monitors starting the bottom of page 62.


***The rest of this post will focus on the Agent-free monitoring.***

Once you install the MP Suite and you have your Agent-free devices discovered, the next thing to do will be to right click on the column headers and click on Personalize View.  Then there will be several options here you can select to see more info about your servers in the main view.

























Check the items you want.  Some may have the same information as others so you may have to play with it a little to see which ones you want and which you don't.  

Depending on whether you enabled the Scalable or Detailed discoveries on the Dell Feature Management Dashboard, some items may or may not be running discoveries already.  Those are described in the charts listed above.

But to see the unit monitors light up, you still have some work to do.  As you can see in the first picture above, the Dell Server Memory Group says Not Monitored.

There are 2 types of Unit Monitors to be aware of (Event and Periodic).
  • Event - triggered by the event that the Dell instrumentation logs in the Windows event log, indicating the health of the corresponding object.
  • Periodic - triggered by a periodic poll configured as Interval Seconds.
Next, go to Authoring, Management Pack Objects, Monitors.  Use the Look For search box and type in "Dell Server".


























Find the Unit Monitors that you chose from the charts listed above, that you want to enable.  For example, we will look at the Dell Server Memory Group.  Find the Dell Server Memory Group\Entity Health\Availability\Dell Server Memory Group.  Right click on it and select Overrides\Override the Monitor\For all object of class: Dell Server Memory Group.














On the Override Properties screen, you have several items you can change.  To enable the item, you need to check the Enabled line and change it to True.  And check the Generates Alert option and change it to True if you want alerts from this component.  Other items I suggest you look at include:

Alert on State - do you want an alert only in critical state or in both warning and critical states.
Alert Priority - depending on the component you may want to increase the priority of the alert.
Alert Severity - again, depending on the component and how important it is you might want to change this too.
Auto-Resolve Alert - do you want the alert to auto-resolve when it sees the condition change back to healthy or do you want to have to acknowledge the alert and close it in OpsMgr.
Interval Seconds - depending on the component and its importnace, I wouldn't change this one too much.  The lower the time, the more polling the OpsMgr server will do of that component, which increases load and traffic.
Log Level - is more for troubleshooting.  If necessary, change this from 0 to 1 and you should get a log at the interval seconds time, in a sub-folder at C:\Windows\Temp.

Once you have your options set, save this to a management pack, but don't use the default management pack.
















Give it some time and possibly hit refresh and then your monitors will light up.  Do this for all of the unit monitors that you want to get alerts for.