PRTG Manual: Monitoring Virtual Environments
A highly flexible IT infrastructure is a common need nowadays and virtualization has become an important pillar of the IT all over the world. Applications in your network might be distributed over many different servers, networks, and locations. They might also be in the cloud, and your computations can take place in datacenters spread over the whole world. So, if your network connection or any other corresponding hardware fails, hundreds of applications might be unavailable—an impact to your daily business processes that should be avoided at any costs.
Because of this, monitoring the physical infrastructure of your datacenter is still a must in times of virtual environments. With the layer of virtualization in addition to your physical equipment, your logical infrastructure also needs a close treatment. PRTG assists you to deal with these advanced requirements and enables you to react proactively to issues before they affect your whole system. With PRTG you can monitor all layers of your IT infrastructure in a comprehensible way so you will reduce issues related to dynamic IT environments significantly.
In general, you can assume that with the layer of virtualization you have to monitor a total of four layers in your IT infrastructure:
- Hardware (Server Racks): Usually you will set up your monitoring in the common way and monitor most of the hardware components in your network with PRTG's SNMP sensors. You are able to gain monitoring data about many different device readings such as CPU load, memory, disk space with this monitoring technology, as well as information about network traffic and bandwidth usage of your routers and switches. It is absolutely mandatory for a working IT environment to monitor all hardware components in order to be alerted if something fails or hardware resources are running out. In addition, you can identify potential bottlenecks affecting your virtualized infrastructure. You can set this monitoring up in the usual way.
- Host Server Hardware: It is essential to explicitly monitor the host hardware of your virtualization solution. If you have issues with your virtual machines (VMs), the origin might be a host hardware failure. You should closely monitor your VM host servers to get alerted if the status is other than "normal". Besides the out of the box hardware sensors, PRTG provides specific sensors for various virtualization host servers; the following monitoring data of your host servers will help you prevent issues in virtualized environments:
- VMware: current reading and health status (via WBEM), a general status as shown in vSphere (via SOAP), and disk space of a VMware data store (via SOAP)
- Hyper-V: host health critical values, deposited pages, network traffic, CPU usage of guests, hypervisor, and in total
- Citrix XenServer: CPU, memory, and network usage, the number of running virtual machines on the host server, and load average
- VMs from the "Outside": The virtual machines run on their particular host servers. PRTG can show you the status of single virtual machines and several of their performance counters. It might be helpful to know which resources a single VM uses and needs, but monitoring single VMs is not advisable in every case because it has a noticeable influence on the overall performance. Often it will be sufficient to monitor only VMs that are critical for your network. If a VM reaches its capacity limits, PRTG can alert you and you can conduct the corresponding resolution steps like enhancing this VM's resources. Indicators for a healthy virtual machine that you can monitor with PRTG out of the box are:
- VMware: CPU and memory usage, disk read and write speed, read and write latency, and network usage
- Hyper-V: CPU usage, disk read and write speed
- Citrix XenServer: CPU usage and free memory
- VMs from the "Inside" (Operating Systems): You can monitor the Windows operating system of a single VM with PRTG's standard WMI sensors, for example. With this technology you can access data of various Windows parameters. Other operating systems like Linux/MacOS can make data available via SSH and SNMP. The status of the operating systems on your VMs can indicate potential issues of the same, just like the operating systems on your physical machines that are important for a reliably working IT infrastructure: You can monitor these with the same attention, depending on your application scenario, but be careful due to performance considerations. Especially many WMI sensors can result in load problems, so monitor only really important systems "from the inside". Furthermore, you do not need to monitor every item multiple times. For example, it might be sufficient to monitor free disk space only from the outside of the actual VM.
To monitor your IT infrastructure, best practice is to set up the monitoring of the hardware layer of your datacenter first in PRTG, especially in order to find potential bottlenecks that might have an impact on your virtual servers. Then you can start monitoring your virtual environment itself. If you use several solutions for virtual hosting, it is also a good idea to group related host servers, their virtual machines, and the operating systems together. The screenshot below will give you an idea about how to organize this.
Grouped Virtual Components
The screenshot above shows you the particular group "Virtual Hosting" of an entire PRTG setup. This is an example how monitoring of virtual environments can look like. The sample group contains several subgroups for the virtualization solutions Citrix "XenServer", Microsoft "Hyper-V", and VMware "vSphere". The vSphere group, for example, has three subgroups: we monitor the vCenter VMs and the vCenter Windows system, the performance of the host server, and the storage system of the host.
In PRTG, set up devices that represent the physical hosts of your virtual machines. For example, for your VMware hosts, add devices that represent the ESXi servers, for Hyper-V add devices that represent your Hyper-V host servers, for Citrix add devices that represent your Xen servers.
Then you can add suitable and expressive sensors to the host server devices. Running the PRTG auto-discovery, many useful sensors will be created automatically. There are several pre-configured host hardware sensors available out of the box in PRTG:
- VMware Host Hardware (WBEM) Sensor: monitors an ESXi server via Web-Based Enterprise Management (WBEM)
- VMware Host Hardware Status (SOAP) Sensor: monitors a VMware host server via Simple Object Access Protocol (SOAP)
- VMware Host Performance (SOAP) Sensor: monitors a VMware host server via Simple Object Access Protocol (SOAP)
- Hyper-V Host Sensor: monitors via Windows Management Instrumentation (WMI) or Windows Performance Counters, as configured in the "Windows Compatibility Options" of the parent device
- Citrix XenServer Host Sensor: monitors via Hypertext Transfer Protocol (HTTP)
These sensor types monitor hardware specific counters to ensure that no hardware issues affect your actual virtual machines. Additional sensor types can monitor the host hardware via SNMP (for example, traffic and custom requests), data storage on ESXi servers via SOAP, as well as sensors for network adapters and storage devices that are connected to a Hyper-V host server.
To monitor your actual virtual machines, add them to your host servers in PRTG. For a better overview, you might want to add another device to PRTG that represents your host server and add sensors for your VM to there. The according sensors for virtual machines will show you the performance of single VMs as well as their usage of resources. This will help you identify VMs with poor performance and react proactively before one or more VMs crash. As mentioned above, you can additionally monitor your particular VMs from the inside (which means the operating systems on your VMs) if necessary. See the sections below for details about particular virtualization solutions.
PRTG's VMware Virtual Machine (SOAP) Sensor monitors VMs on a VMware host server via Simple Object Access Protocol (SOAP). With the VMware system, the general idea is to add a vCenter server as a device to PRTG and use the vCenter as parent device where you add the sensors to for your virtual machines. So, in the case of vMotion when your VMs change their host server, PRTG will be able to follow these movements and will never lose the monitored VMs.
For this sensor type, you need the Microsoft .NET Framework version 4.5 or later running on the probe machine. If you use many VMware sensors, we also recommend adjusting the settings on your VMware host server to accept more incoming connections.
This screenshot a sample vSphere group in PRTG. As recommended, the VMware virtual machines are added to the vCenter device. There is also a dedicated device for the vCenter Windows operating system with common WMI sensors for CPU, memory, disk, and network usage. The ESXi host servers are organized in their own groups regarding performance and storage. In this example, PRTG monitors the hosts with the standard SNMP hardware sensors as well as with the specific VMware ESXi host sensors.
PRTG's Hyper-V Virtual Machine Sensor monitors VMs via Windows Management Instrumentation (WMI) or Windows Performance Counters, as configured in the "Windows Compatibility Options" of the parent device. With this hybrid approach, the sensor first tries to query data via performance counters and uses WMI as a fallback if there are no performance counters available. Performance counters in general need less system resources than WMI. We recommend using System Center Virtual Machine Manager (SCVMM) as parent device for this sensor type, because when your VMs change their physical host with Live Migration, this sensor will still be able to continue monitoring. You should also disable User Account Control (UAC) in the Windows operating system of the VM.
This screenshot shows a sample Hyper-V group in PRTG. There is a dedicated group for failover clustering where two nodes are monitored with several SNMP and WMI sensors, as well as Hyper-V Host Server sensors and sensors for the Hyper-V virtual machines. This ensures that Hyper-V and failover clustering works without any issues. The Hyper-V hosts are monitored the same way, organized in a dedicated group for hosts.
Note that we recommend adding the particular virtual machines to the SCVMM server if possible! We pointed out above that you should add the VMs to an SCVMM server to avoid issues with Live Migration. You can see this in the group "System Center". The particular VMs are added to the dedicated device "Hyper-V Guests".
PRTG's Citrix XenServer Virtual Machine Sensor monitors VMs via Hypertext Transfer Protocol (HTTP). For this sensor type, you have to add a device to PRTG that represents a Citrix XenServer with version 5.0 or later. Another requirement is the Microsoft .NET Framework: You have to run .NET version 4.5 or later on the probe machine where you add this sensor to.
In a XenServer pool, every host knows each running VM. Because of this, there is no central instance that provides all available data so it does not matter on which host you query your VMs. All queries on any host are automatically forwarded to the pool master, which manages the XenServer pool. So it is sufficient to create the desired sensors for your XenServer VMs on a device that represents one host server of your pool. PRTG's XenServer sensors can figure out by themselves which host is running and retrieve the according data.
This screenshot shows a sample XenServer group in PRTG. There are two devices for XenServer hosts (Xen 1 and Xen 2), each with a Citrix XenServer Host sensor and several Citrix XenServer Virtual Machine sensors for the particular VMs on this host. Furthermore, the Windows operating system is represented as a dedicated device ("virtualcontrol") that is monitored with several WMI sensors regarding CPU, disk, memory, and network usage.
The sensor types described in this section monitor virtual machine specific counters to ensure that all your VMs have enough resources available. If a VM is overloaded, PRTG can notify you immediately and you can proactively take care of issues before a particular VM has an outage or other failures. Additionally, we have shown an idea for a structured virtual monitoring with several recommendations.
You can find all available sensors for virtual servers and the according virtual machines in section List of Available Sensor Types—Virtual Servers Sensors.
For best performance when monitoring virtual environments, we strongly recommend that you use a computer with Windows Server 2012 R2 to run the PRTG probe with the according sensors. So for example, you can run up to 300 VMware sensors in a 60 seconds scanning interval on Windows 2012 R2, while you can only use 30 VMware sensors with the same interval on Windows 2008 R2.
Knowledge Base: I run PRTG on VMware. How can I obtain best performance?