Managing changes in your virtual machines is like playing whack-a-mole. You probably recognize these questions:
- Did anybody change anything on that VM?
- Are VMs running worse because of a change we made? Are they running better?
- If we make a change to this VM, what else in our environment will be affected?
In my last post, I described the new Getting Started dashboard in Foglight for Virtualization, Enterprise Edition 8.0 (FVE 8.0) and several of the questions it can answer for virtualization administrators. In this post I’ll cover the new Change Analyzer feature in FVE 8.0 that helps you answer questions about changes to your VMs and take actions right away.
Things happen to VMs. I don’t know why, they just do.
Suddenly, a VM isn’t working properly and you ask, “Did anybody change anything on it?” and everyone says, “No.” So you spin your wheels figuring out why performance has dropped off or why the VM crashed. Eventually you try to put things back the way they were before.
With Change Analyzer, you don’t have to ask whether anybody changed anything — you just know. It’s like an audit trail for virtual machines, as you can see in the screenshot. You know what changed and who changed it. For that matter, you can even see if it wasn’t a “who” but a “what” — like VMware’s Distributed Resource Scheduler (DRS) or some other software tool — and you can still find the change, even among tens of thousands of VMs.
Best of all, you can fix most mistakes introduced by users. Change Analyzer is more than just a reporting tool; it lets you undo accidental or undesired user changes in your virtual infrastructure. It hooks into APIs recently exposed in vCenter and aggregates data to show you in real time all the changes that occur.
Are things running poorly because we made a change?
Did you ever make a change, like powering a VM on or changing its resource pool, and suddenly get slammed with support calls about degraded performance? ChangeAnalyzer lets you correlate changes in performance to changes in your infrastructure.
Suppose you make a configuration change. Using a health baseline, Change Analyzer compares performance after the change against performance before the change and shows you the delta as a percentage. You can search for all VMs whose performance has degraded by a given percentage or more; then see which changes are associated with the drop.
It works in the other direction, too. You can determine the changes that boosted performance by a given percentage and implement them elsewhere for better overall performance.
If I make a change on this VM, what will that affect?
Remember the cartoon gag where the character closes one drawer and another one opens? Then he closes the second one and a third one opens? VM changes can be like that, so we’ve built Impact Analysis into Change Analyzer. For any host with multiple VMs, you can now anticipate how changes to one will likely affect the performance and capacity of others.
Impact Analysis covers the most common and far-reaching changes you make in your virtual environment:
- Change VM configuration
- Change resource pool
- Move a VM
- Power a VM on/off
- Change high-availability failure configuration
- Deploy a VM
- Migrate a VM
Before you make the change, you can now see what it will most likely affect, like the host machine, HA status and neighboring VMs. Change Analyzer estimates impact based on data it collects from vCenter and on relationships FVE maps out across your virtual infrastructure.
You can find the answers to some of your virtualization questions manually, but you know what a drag it is poking around in one VM after another, trying to see what’s healthy, what has changed and what affects performance. FVE 8.0 is the first tool to give you such a broad overview of your virtual environment AND let you do something about it.