VMware vRealize Operations (vROps) 7.0 uses Badges as a way to evaluate objects. The three major badges it uses are Health, Risk, and Efficiency. As I’ve written previously, every VMware administrator needs to understand what comprises these badges, and how they are applied to the objects discovered within your stack. Badges summarize a lot of information, and help you take efficient action to keep your customers happy. Knowing what action to take requires understanding Alerting in vROps.
Badge values are based on scores calculated from the Alerts against them. VMware documents the thresholds for each of the major badges. For the Health Badge, an overall score is between 1-100, where 100 is good/green and 0 is bad/red. If the Health Badge for a particular object is unknown it will be -1 and will show as unknown/grey. The score, based on a proprietary algorithm within vROps, is derived from the criticality of Alerts in the Health category. As a reminder, Alerts can be Critical, Immediate, Warning, or Info.
I generally think of the Health Badge as showing the current state of an object, the Risk Badge as showing “tomorrow’s state” of an object, and the Efficiency Badge as showing “next week’s” state of an object. Users want to take action on Health related items immediately, Risk related items “tomorrow”, and Efficiency related items “when you have time”.
The Risk Badge is the overall score for Risk, items that you will want to take care of “tomorrow”. The final score is between 1-100, where 100 is good/green and 0 is bad/red. If the Risk Badge for an object is not known it will be -1 and will present as unknown/grey. The score, is derived from the criticality of Alerts in the Risk category. Keep in mind, Alerts can be Critical, Immediate, Warning, and Info severities.
Finally, the Efficiency Badge is the score for Efficiency, items that you will want to address “when you have time”. The final score is between 1-100, where 100 is good/green and 0 is bad/red. If the Efficiency Badge for an object is not known, it will be -1 and will show as unknown/grey. The score, is based on the criticality of Alerts in the Efficiency category.
Here, I’ve created a dashboard allowing the user to select a VM and see that VM’s Health, Risk, Efficiency, and related Alerts (this dashboard is available at VMware {code}). This same dashboard can be created for any object you want—for example, a NetApp array, an EC2 instance or MS SQL Server database.
Badges
I think of Badges as buckets; Alerts go into the buckets, and as the buckets fill up, their Badge scores change. To empty the buckets, the administrator needs to remove Alerts. To remove Alerts, you need to fix the problem causing the Alert.
It’s important to note that VMware vROps has three important constructs: Alerts, Events, and Alarms. I polled Twitter and received some great feedback from VMWare’s own John Dias (@johnddias) on these. John indicates: Events are “things that happen” while Alerts are making you aware of “bad things that happen” which could be Events but don’t have to. Here is how I define each:
Alerts
An Alert is a vROps construct making the user aware that something is happening, i.e. CPU is being consumed, Memory usage is high, etc. There are four Alert severities: Critical, Immediate, Warning, and Info. Alerts are presented within vROps via the Alerts tab, the Alert List widget or the Top Alerts widget. Here is the Alerts tab from my vROps 7.0 instance:
Here are the Top Alerts and Alert List widgets as shown for a selected Datacenter:
Events
An Event is a vROps construct indicating something has happened. Events can trigger Alerts, but don’t have to. An example of an Event would be something like a disk filling up or a VM rebooting. Events can be found via the Events tab from the object detail listing within vROps. Here is the Events tab for a VMware Datacenter:
Alarms
Alarms are synonymous with Notifications, things like emails, text messages, instant messages, and other forms of notification. Alarms are how you are made aware of Alerts in your vROps environment.
My Twitter Poll also generated feedback regarding logs and their relationship to Alerts. Michael Ryom (@michaelryom) responded to my poll indicating: “…logs are things that are happening…”, meaning they precede Events and Alerts. Perhaps this is our next blog.
Use Badges to quickly assess the state of your environment and Alerts to guide your correction efforts. If you’d like to see how Blue Medora can help you see the entirety of your stack in vROps, ask us for a guided demo, or try it yourself.
