Thursday, April 23, 2015

Part 13: The Power of Alert Definitions in vRealize Operations Manager 6.0!

With the release of vRealize Operations Manager 6.0, I have started this series which gives you a scoop on how does this new release looks like and works. In my humble opinion this release of the product is way more mature from the previous release as it gives you a plethora of options around capacity and performance management which were not available with the previous release.

While the traditional monitoring softwares look at Key Performance Indicators and always work with the defined thresholds, vCenter Operations Manager (previous release) gave a new meaning to the world of monitoring by using the concept of Super-Metrics. In simpler terms, a supermetric is a value which is derived by using a mathematical operations on one or more metrics which are collected by vCOps/vROps. This results into a value on the basis of which it becomes easier to take decisions. These decisions can be around troubleshooting performance issues, capacity augmentation or capacity reclamation. 


Some of these Super Metrics are represented by using standard names and badges which can clearly help you identify the problem areas of your environment. The three major super metrics which stand out are HEALH, RISK & EFFICIENCY. While Health, Risk & Efficiency were calculated by using the minor badges in vCOps 5.x, with vROps 6.0 these badges are defined through Alerts which are generated by vROps. This is a major change from vCOps 5.x and this explains why the score are no longer available on these major badges. This also explains the default dashboard which shows us when you login to the vROps product UI. This is called the Recommendation Dashboard and you can see alerts in the lower pane and the Health, Risk and Efficiency badges on the top. Let's have a look at a screenshot to explain this:



As we know from one of my previous post the scores on the main screen on the minor badges are no longer visible. The primary reason behind this change is because of the simple fact that the minor badges now do not make up the major badges of Health, Risk & Efficiency.

While the screenshot above is from the recommendations dashboard, you would need to change the alert definitions if you wish you impact these badges by various alerts which are defined by the system or are user defined. To do so, you need to go into Content -> Alert Definitions as per the screenshot below:



Here you can edit the alert definition and define the impact it can make on a particular badge associated with the specific object or an object type.

I remember that someone told me that the badge scores were good and they should have remained with vROps 6.0. In my humble opinion badge scores are still there but they are gone from the default dashboard which is called the "Recommendations" Dashboard. The function of this dashboard is to provide you some action items based on the metrics which are changing in your environment and the best way to provide these recommendations would be in the form of the alerts which at the end of the day help you monitor your key performance indicators. Was that not the reason why you invested in a solution like this???

Would be happy to know your thoughts on this, hence please feel free to share via the comments section... Stay tuned for more coming on vXpress...

Till then!!


Share & Spread the Knowledge!!



4 comments:

  1. I would like to know if need to create an alert to call vRO to add more resources in the virtual machinal (CPU, by example).

    ReplyDelete
    Replies
    1. Yes that is correct. That would be an action associated with an alert which would automatically call a VRO workflow to do a rememdiation task

      Delete
  2. Hello Sir,

    Kindly Help us on following

    we created vm power off alerts with following condition

    "Change in Condition :- If VM is Powered Off and The Host is Connected alert VM is Power Off.

    We have also tested the Alert with Multiple Condition and all the test have Passed. "

    But from last few dayswe are getting VM Power OFF alert from vROPs after Host getting host disconnect from vCenter Inventory .

    Regds,
    Mitra.

    ReplyDelete
    Replies
    1. You do not have to use the host disconnect here... You can use a simple metric to create a symptom which monitors the power state of a VM.

      System|Powered On = 0
      Wait cycle =1 and Cancel Cycle =1

      Just with that single symptom you can monitor VM power state.

      Delete