July last year I wrote a blog post called "Oversized and undersized VMs plus one IT geek meltdown" which summarized to say that the reports which Veeam has for oversized and undersized VM’s are great but I run Hyper-V and I want these same types of report for my Hyper-V systems. The community has had a lot of feedback on this and there have been several follow-up blog posts which have helped Operations Manager administrators to effectively assess their Hyper-V systems.
Fast forward nine months and it looks like Veeam has either read my blog or I was lucky enough to channel what their development direction would include (most likely a little of both). I was invited to be a tester on the upcoming Veeam Hyper-V management pack and I am impressed with what I’ve seen so far. So… what’s in the (currently under development) Veeam Hyper-V management pack?
From a high level this blog post will discuss the installation steps required for this management pack, views included in the management pack, reports in the management pack and an extra-special extra. Please note that all of this is still under development and subject to change.
Installation was a breeze – add the management packs and make sure that proxy is enabled for the various hosts. There were no additional components to install, and no other configuration was required.
Views in the Management Pack:
Views include both standard views (alerts, state, performance and topology) as well as Operations Manager 2012 specific dashboard views.
Alert views covered the alerts identified by the management pack. For my environment this included one critical alert and two warnings alerts. The first warning was related to an old checkpoint which existed on one of my virtuals.
The critical alert was related to a service which was failed on one of my domain controllers as shown below.
The second warning alert was the most interesting one which indicated that there are memory pressure issues with one of my virtuals. I used Hyper-V manager to change the memory configuration on this virtual which resolved the underlying issue and closed the alert triggered by this monitor.
State views covered any clusters in the environment as well as any hosts and virtual machines.
The host view provides quick insights in the health of the Hyper-V hosts, underlying virtuals and gives details about the hardware on the host.
The VM’s view provides a list including many of the key pieces of information for virtual machines as shown below (plus other fields including CPU priority, CPU reservation, disk allocation, version, generation, virtual NUMA spanning, storage, network and more).
Performance views are available for a variety of items monitored by this management pack. The example below shows a host metric around the average memory pressure on the various Hyper-V hosts.
Topology views are available for both the Computer Topology (shown below) and the Storage Topology.
What would a new Operations Manager 2012 management pack be without some new dashboards and some new widgets? The new Veeam Hyper-V management pack includes some excellent dashboards including the items shown below for top hosts and top virtual machines.
Note how items which are past warning thresholds change color such as the DB02 example which is in a warning threshold. These thresholds can be personalized so that you can determine what thresholds you want for your environment.
Reports in the Management Pack:
Ok, that’s all cool, but what about the reports? That’s what started this discussion originally. Did Veeam create new reports which cover the requirements?
The list of reports currently available is shown below with sample outputs.
The Configuration Tracking and Alert Correlation report provides information as to what is alerting and changing in this Hyper-V environment as shown below.
And the "Virtual Machines, Idle VMs" report is easy to identify what is idle across these hosts.
What about the oversized and undersized reports? The oversized and undersized reports exist and they look a whole lot like the experience that we get within the Veeam management pack. Subsets of the oversized VM memory and CPU as shown below.
And the equivalent reports also exist for undersized VM memory and CPU as shown below:
Tasks in the Management Pack: (IE: something extra-special)
If you have worked with Hyper-V you may have noticed that once you install Hyper-V on a system, task manager isn’t as useful as it is without Hyper-V installed. We can see
In the example below for the processes tab I can see that the memory is being used by I can’t tell what is using the memory (note that the largest memory usage is 132.9 MB for a 48 GB system).
Veeam included a Hyper-V task which is targeted at the Hyper-V hosts.
Their task based version of this type of a view connects and gathers real time performance information from the host but it provides it in a way that it’s easy to identify what Hyper-V guests are using what amount of CPU and Memory resources on the Hyper-V host.
The Veeam Hyper-V Task Manager is an excellent addition to the tool belt and is something that I see myself using regularly in my Hyper-V labs.
Summary: All I can say is that when this comes out – it will be a good day to be a Hyper-V user! I’m sure that more information will soon be available once this product is released at http://www.veeam.com.