Having heard some great things about the new vRealize Operations Manager 6, I installed it in my lab to get familiar with the product. At first glance, while feeling familiar, there is no doubt that VMware has really polished this product. From an easier deployment model, to the welcome addition of new vSphere Dashboards that actually are meaningful, to the overall feel of the product, it just feels, well, like a finished product should feel.
While there’s still that familiar look to the home landing page, you can’t help but get a sense that VMware took some advice and decluttered. Some of the constant complaints I seen from my customers with vCOPs was the sheer amount of information on the main vSphere Dashboard. While info is important, it’s more in how and why the information is presented that’s even more important. Users want an operational mgmt tool where they can navigate easily and see relevant information front and center and not have to spend additional time digging through a plethora of information and heatmaps, especially in larger environments.
While this isn’t a primer on vRealize Operations Manager 6, I wanted to focus on something that I found while playing with the product in my Home Lab. I call it Actionable Alerting, though I’m sure VMware has a pithy name for it. As you can see in the screenshot above in the lower lefthand box under Health, vROPs is reporting Health Alerts in my MGMT Cluster, 3 to be precise.
At the very front page, i’m told there is unexpected CPU workload and 3 objects are impacted. I’m also told that vROPs is providing 2 recommendations. Let’s go take a look at this shall we?
Let’s focus on the second one as the other two are related to the MGMT suite of products and the second one is my vCenter Appliance. I am running vSphere 6 Beta in my lab so I’m only going to focus on the vROPs interface as to not break any NDA. Let’s click on the Second Alert and see what information if presented:
Before I move on, I want to just take a moment and focus on the information provided in this screen. If you have used the previous versions of vCenter Operations Manager, you would be provided with tons of information which again, you would have to sift through and hopefully find specifics around the issue. With vROPs 6, you can see the information is relevant, and presented in a neat and orderly fashion. Again, much better from a usability standpoint.
Now let’s focus on the recommendations. We see that the CPU usage on my vCenter Server Appliance is or has run outside of normal and has exceeded the threshold, in this case 70, which has been reported as an anomaly. You can see more detail and graphing by selecting the to the left of the alert.
So, what does vROPs recommend to address the issue? The initial recommendation is to check the guest applications to see if this is an expected behavior. Let’s say you check and you see that CPU is and has been reaching these levels for a while and it is determined that adding additional vCPU’s will address the issue. Well, more than likely you would just go back to your WebClient, and if Hot Add is enabled, add vCPU’s (of course keeping within NUMA boundries) but do we really need to go to another interface to address this. The answer is not necessarily. Let’s click on other recommendations.
As you can see to the right, vRealize Operations Manager is giving you the option to address the issue without having to leave the interface. As vSphere Admins, you may ask “is this functionality something I would leverage?” Maybe, maybe not, but let’s look at this from an Application owner. Remember, VMware is helping us achieve the self service model and if I’m an application owner that has access to vROPs looking at my application and I wouldn’t have access to the WebClient or any other Infrastructure Administration Tools, this could (I stress could) be beneficial too me.
Let’s see what options we have when we click the tab.
vROPs provides a recommended vCPU setting that it thinks will address the Anomaly. You can also Power off the Guest if it’s required to add vCPU (hot add is disabled) and even take a Snapshot prior to the change for easy rollback. For informational value I changed the vCPU’s to 6 on my vCenter Server Appliance and was able to watch what occurred from the Recent Tasks view.
As you can see this action failed because the vcsa doesn’t have hot add vCPU capability and required a reboot.
This is a pretty nifty feature that’s built in to the vROP’s appliance. Nothing outside of the appliance had to be installed or configured. I can see where this can be a powerful feature and allows for true Actionable Intelligence for Alerts.
I recommend you take vROPs for a spin. VMware has done a fantastic job in polishing this entire product. I look forward to work with it more with customers and hear their feedback.
As always, your feedback is welcome!