Monitoring networking infrastructure is the best way to ensure that all is running smoothly at all times. It is also one of the most important tasks of network and systems administrators. There are various ways monitoring can be done. First, it can be a manual process. While this may work—especially in smaller environments—it is very time-consuming and tedious, often requiring a dedicated resource. This is why automated monitoring is much more popular. And when talking about automated monitoring, there are basically two ways it can be done: agent-based and agentless. Today, we’re having a look at some of the best agentless monitoring tools.
We’ll start off our discussion by having a look at monitoring in general. We’ll do our best to explain not what it is but what we feel it should be. We’ll then discuss the differences between agent-based and agentless monitoring. We’ll have a look at the pros and cons of each. After that, we’ll explain SNMP monitoring. It is important as it is the primary agentless monitoring scheme. But SNMP is not the only monitoring system and we’ll, therefore, discuss other monitoring protocols that are available. And once we know all there is to know about agentless monitoring, we’ll be ready to review the best agentless infrastructure monitoring tools.
About Infrastructure Monitoring Tools
Infrastructure monitoring can be defined as the process of ensuring that critical operational parameters of IT infrastructures remain in their normal range at all times. Unfortunately, this definition is a bit vague. It fails to explain what critical operational parameters are and what their normal range is. There’s a good reason for that, though: it varies. For instance, while CPU load might be a very important parameter on servers, it’s not so much on network switches. The best monitoring tools will automatically pick the right parameters to monitor for different types of infrastructure components.
Another important aspect of monitoring tools is the way they communicate information. It is customary for such tools to have some type of a dashboard where the primary parameters are either displayed as gauges or as graphs showing their evolution over time. We also often see tools that will have a list of all devices on their primary screen and allow you to see individual device parameters by clicking the devices in the list.
And since you possibly have better things to do than monitor a monitoring tool dashboard all day, most tools also have some form of alerting that is triggered whenever a monitored parameter exceeds its normal range. Some tools have pre-built alerts while others allow you to define them at will.
Agent-based Vs Agentless Monitoring – The Differences
There are multiple ways that infrastructure monitoring tools can get operational parameters from the devices they monitor. All methods fall into one of two categories: agent-based or agentless. As the name implies, agent-based monitoring uses an agent that’s running on the monitored system. That agent is a piece of software that collects operational data and then sends it to the monitoring system. Agentless monitoring relies, instead, on the built-in capabilities of monitored devices to read their operational parameters. Each method has its advantages and disadvantages. Let’s see what some of them are.
- It gives monitoring system developers the best flexibility in defining what information is exchanged between the monitoring tool and the monitoring system.
- It will often provide much more information than what’s available using standard agentless protocols.
- Since the collection of the parameters is done by the agent, the monitoring tool has less work to do. It makes for a simpler tool.
- Support of new device types doesn’t require updating the tool. It’s just a matter of installing a new agent.
- What can be monitored is limited by what agents are available.
- Agents running on them add an extra load (in CPU, memory, and disk space) to the monitored systems. This can potentially affect performance.
- Not all systems permit the installation of third-party agents. Networking devices, for instance, rarely do.
- There’s no extra software to install on the monitored systems, making setup and deployment much easier and faster.
- The monitoring is always done the same way regardless of the monitored system.
- There could be a limitation to what can actually be monitored. Some parameters simply can’t be remotely monitored.
- The security of some monitoring protocols leaves a lot to be desired.
- More work is required from the monitoring tool.
SNMP To The Rescue Of Agentless Monitoring
Many agentless infrastructure monitoring tools rely on the Simple Network Management Protocol, or SNMP, to accomplish their feat. SNMP specifies a communication scheme between a monitoring platform and the monitored device as well as the structure of the monitored data. Different SNMP-enabled devices make various operational parameters available via the SNMP protocol. SNMP monitoring tools connect to those devices at regular intervals and read their operational parameters. For example, CPU utilization or memory usage can be read through SNMP.
For network equipment monitoring, some counters called bytes in and bytes out can also be read via SNMP. Typical network equipment has a pair of those counters for each network interface. These are not gauges, though like the CPU and memory usage gauges discussed above. They are counters that indicate how many bytes have been transmitted and received and they increment each time a packet is sent or received. By using a known polling interval, simple mathematics will allow monitoring systems to calculate average bandwidth utilization.
SNMP Is Not The Only Monitoring Protocol
As good as SNMP is, it’s not perfect and not all parameters can be monitored using the protocol. For instance, while servers often have SNMP—although it may not be there by default and might need to be installed—many of their advanced operational data is not available to SNMP.
Other techniques and protocols can often be used be used. In the Windows world, the WMI (Windows Management Instrumentation) protocol can be used as a mean of communications between monitoring systems and monitored servers. Other protocols such as WBEM (Web-Based Enterprise Management) or CIM (Common Information Model) are used in a similar way in the non-Windows world.
But if we need to install additional software components such as SNMP, how is it different from using an agent? The main difference lies in the fact that the agent is from a third party while the SNMP module is an operating system extension.
Our Top Agentless Monitoring Tools
So, if you’ve decided that you’d rather not install third-party agents on your systems and want to stick with agentless monitoring tools, here are short reviews of the best packages we could find. All of them use SNMP to monitor remote systems and some also allow other protocols to be used. The different tools on the list are not really ranked and don’t assume that the first ones are necessarily better.
SolarWinds has become a household name with network administrators. The 20-year old company makes some of the best monitoring tools out there. And it also makes a wide range of free tools that address specific needs of network administrators such as an excellent subnet calculator or a TFTP server.
SolarWinds’ flagship product is called the Network Performance Monitor, or NPM, a complete network monitoring solution. The tool’s most important characteristics are its simplicity, its scalability, and its customizability. The product’s graphical user interface is very intuitive and easy to use and learn. The NPM will scale from very small networks to huge installations with thousands of devices over multiple sites. As for the product’s customizability, it is everywhere: in the dashboard, the views, the reports, the alerts.
The Network Performance Monitor’s alerting system is one of the best available. It can either be fully customized to your needs or be used out-of-the-box with little or no configuration. It has built-in intelligence and will not, for instance, send notifications for “unimportant” events in the middle of the night or send hundreds of “device unreachable” notifications for a device located behind a down router or network switch.
Prices for the SolarWinds Network Performance Monitor starts at $2 955 and goes up according to the number of devices to monitor. If you want to try the product before buying it, a free 30-day trial version is available for download from the SolarWinds website.
As good as the SolarWinds Network Performance Monitor is, it’s specifically made to monitor network devices although it could monitor any SNMP-enabled device, even servers. But if what you need to monitor are servers, their processes, and the applications running on them, what you need is the SolarWinds Server and Application Monitor. As you’d guess from its name, this tool monitors applications as well as the server hosting them. The software scales from very small networks to large ones with hundreds of servers–physical or virtual–spread over multiple sites. It can also be used to monitor cloud-hosted services like those from Amazon Web Services and Microsoft Azure.
Setting up the tool is easy and the initial configuration is done with the help of an auto-discovery process. It is a two-pass process that will first discover servers, then the applications running on them. To speed up the process, a list of applications to look for can be supplied. After the initial setup, the user-friendly graphical user interface makes using the Server and Application Monitor easy. All the collected data about your servers, processes, and applications can be displayed in either a table or graphic format.
Prices for the SolarWinds Server and Application Monitor start as $2 955, just like the NPM and it too is priced based on the number of monitored servers. And like its cousin, a free 30-day trial version is available for download.
The Paessler Router Traffic Grapher, or PRTG, is an integrated tool that, despite its somewhat misleading name, can monitor any system, device, traffic, and application in your IT infrastructure. PRTG is quicker and easier to install than most other network monitoring tools. In fact, Paessler, claims you could start monitoring within two minutes. The tool’s auto-discovery system will scan network segments and automatically recognize a wide range of devices and systems. It will then create sensors from predefined device templates.
The platform also offers a very flexible and customizable alerting system. You can even get alert notifications pushed to your mobile device when using the free client apps for Android, iOS, and Windows Phone. You can also get them via more standard means such as email or SMS, according to your needs.
PRTG is available in a free, full-featured version. It is, however, limited to 100 “sensors”. In PRTG parlance, any monitored parameter counts as one sensor. For example, monitoring the bandwidth on each port of a 48-port switch will count as 48 sensors. To monitor more than 100 sensors, you’ll need to purchase a license. Price increases with the number of sensors and starts at $1 600 for 500 sensors up to $14 500 for unlimited sensors. A free device-unlimited 30-day trial version is available.
4. WhatsUp Gold
WhatsUp Gold was well-known as an up-or-down type of monitoring tool for ages. Luckily, the product has evolved into a full-fledged monitoring system with all the bells and whistles. It has—and has had for its entire existence—one of the best alerting systems and it can be configured to transmit alerts using a multitude of ways including email and SMS, to name a few.
WhatsUp Gold won’t only monitor network devices, it can also monitor servers and selected services and processes. It can, for example, monitor Exchange and SQL servers, Active Directory, IIS and Apache Web services. And if your organization is using cloud-based equipment, WhatsUp Gold will also monitor AWS or Azure installations.
Pricing information for WhatUp Gold can be obtained by contacting Ipswitch’s sales or one of their many resellers. A free 30-day trial version can be downloaded.
5. Nagios (Core and XI)
Two versions of Nagios are available. There’s a free open-source version called Nagios core and a commercial product called Nagios XI. Both use the same core–hence the name of the free version–but while the free one relies on community-developed add-ons and plugins for most monitoring tasks, the commercial product includes them.
Nagios a great monitoring solution for applications, services, operating systems, network protocols, systems metrics, and network infrastructure. And if you need more than what the product can do out of the box, third-party add-ons will let you monitor virtually anything.
Nagios XI is available in Standard and Enterprise editions. The Enterprise Edition has some additional functionality and includes features to assist in large-scale configuration, forecasting, and scheduled reporting. Prices for the standard edition start at $1 995 while the Enterprise edition will cost you at least $3 495. A free version of Nagios XI, which you can use as a trial, is available but it is limited to monitoring seven devices.
6. ManageEngine OpManager
ManageEngine is another well-known publisher of network administration tools. And the ManageEngine OpManager is an all-in-one package that will monitor the vital signs of your servers (physical and virtual) as well as your network equipment and alert you as soon as something is out of specs. The tool features an intuitive user interface that will let you easily find the information you need. The product also features an excellent reporting engine along with some pre-built as well as custom reports. To complete the package, this system’s alerting features are also very complete.
The ManageEngine OpManager is available in two versions, The Essential edition is targeting small and medium organizations with up to a thousand devices with prices starting at around $700 for 25 devices. For larger organizations, there is an Enterprise edition which can scale up to ten thousand devices. Its price starts at under $20 000 for 500 devices. Like with most full-featured commercial monitoring tools, a free 30-day trial is available.
7. OP5 Monitor
OP5 Monitor is an open-source (although not free) agentless monitoring tool which is based on a Naemon, a fork of Nagios. The product claims to be “the enterprise-level open-source monitoring solution“. This tool is loaded with excellent features. There is, for instance, as a fully customizable dashboard where you can display what’s important to you. The tool also scales very well, particularly in distributed environments.
Alerting is another strength of OP5 Monitor. Alerts don’t only trigger notifications, they can also launch event handlers: scripts that can make issues self-healing. The system integrates easily with other systems–such as ticket management or CRM–and it has a developer-friendly API if you want to extend its functionality.
To get started with OP5 Monitor, there is a free 30-day trial download. which is ideal for those who want to get hands-on straight away. Alternatively, if you just want to get a feel for OP5 Monitor and its features, there is a demo environment. Pricing information can be obtained by contacting OP5 sales.
Last on our list is Zabbix, one of the best free and open-source system monitoring platform. This enterprise-grade system can scale from small to very big networks. This system can monitor networks, both local and cloud-based servers, and the services running on them without requiring an agent.
Zabbix is different from other free and open-source products. While the product itself is free, ancillary services are sold by Zabbix. Among the services you can purchase are five levels of technical support and a complete certification training program. This is different from many open-source products which only offer community-based support which, by the way, is also available for free and very good. Finally, its alerting and reporting are two areas where this product shines.
Besides the high price tag, Zabbix has all you’d expect from an enterprise-grade monitoring tool. The only investment you need to give Zabbix a try is your time.