Last week I was fortunate enough to attend a briefing from VMware on new upcoming releases of our beloved vRealize Operations, and today I can talk about them! vROps has really advanced tremendously in the last years and it is now for me the solution for monitoring in a VMware multicloud environment. The platform is well positioned for HCI operations (think vSAN, vxRail), hybrid cloud monitoring, app-aware operations, compliance use cases and the aforementioned multicloud.

vROps Cloud has been announced today. All SaaS offerings are now simply named ‘cloud’, so Log Intelligence will become Log Insight Cloud. Watch for other announcements this VMworld…. There will be a tech preview in Octobre – around VMworld EMEA I guess – and a beta early next year for launch around mid next year. It will require simple installation of a proxy in your network and registering a vCenter. VMware on AWS will be even simpler.

vROps v8 will also be available around Octobre. There are a lot of improvements in Workload optimisation and cost to include cloud services such as Amazon. The AWS and Azure management packs will now come with the product out of the box. I think there is a blog post necessary only to talk cost or workload optimisation. Let me concentrate here on a few other remarkable new features in intelligent remediation. Still my favourite use for vROps…

Troubleshooting workbench.

I will try to avoid some popular marketing terms here, but it is hard to do since this is an incredible new feature, with some stuff that learns … (get it ?) and one dashboard with all info to quickly troubleshoot a problem from app to infra…It gathers and correlates info on events, anomalous metrics and property changes. You can change time and topology to find out more and you can run multiple workbenches at once.

troubleshooting workbench

Application monitoring.

The new application monitoring using Telegraf agents is extended with 20 new applications and services! NTPD, Java, and Websphere are now supported and you can write custom monitoring scripts. These options bring vROps more and more in the APM space, along with the following novelty!

Native Service Discovery.

vROps now has built-in agent-less service discovery for 41 known services through vmtools! You can also add your own services to a white list. It will build application relationships and you can run agent-less VM actions for troubleshooting, like getting top-n processes.

Continuous availability.

vROps can now go beyond HA and replicate between sites over a layer3 network connection. So you can have a site failure and still access vROps. I know this is an important feature for large corporations with worldwide datacenters.

All in all vROps 8 and the SaaS offering offer some great new features. If you are a vSphere customer and are moving towards SDDC on prem and cloud offerings such as AWS, you would be well advised to take a closer look!

This is a little view that I created in vRealize Operations because of a customer issue. As you hopefully know in vROps 7.5 the idle VM detections is back. You can switch it on in the reclaim settings. On the home screen go to the Reclaim dashboard and click the wheel next to the title. Now you can switch on idle detection, powered-off VMs and snapshots. Even orphaned disks can be hunted down!

The idle detection looks if CPU consumption in the chosen period is less than 100 MHz. The customer wanted to check also memory, disk and network access to decide on which VMs needed to be singled out for deletion.

I created a view using the new colorisation in v7.5. Attention! Colors only work in dashboards you create with this view or the view itself. In reports they do not show up.

I put in some specific parameters, but you can change them. CPU and memory color red if they use over 500 MHz or 500 MB respectively. Disk and network access go red over 100 kBps. CPU, memory, disk and network are sorted in ascending order.

The preview is on vSphere world. If you get VMs that are green for all metrics over a month, you know they are good candidates to delete… In the image above you see a preview in my little home lab…

We can now use this view in a dashboard so that we can click on a VM and get some more info such as the parent host and cluster and data stores. Well any excuse to use the exquisite new advanced object relations widget! The sparkline widget is used with the VM Capacity metrics to show the configuration of the VM. Of course you can build your own dashboard on top of this!

To use this in your environment download the zip file from VMware Code and unzip. First install the view by clicking on Dashboards, Views, Manage Views and import “historic idle detection view.zip”. Afterwards go to Manage Dashboards and import “historic idle VMs dashboard.zip” there. Both View and dashboard are now available in your environment.

Let me know what you think!