Dynatrace extends Synthetic Monitoring capabilities with Network Availability Monitors to validate the availability of infrastructure and services

Dynatrace extends Synthetic Monitoring scope to your tech stack's network and transport layers by introducing new synthetic test types: ICMP Ping, TCP port check, and DNS-all under Network Availability Monitoring (NAM). Your teams get 24/7 insight into the health and availability of your infrastructure and network. Synthetic Monitoring also now provides basic information about the health of network services that aren't based on HTTP(s) and, thus, were not targeted for synthetic testing in the past.

Current synthetic capabilities

Dynatrace Synthetic Monitoring is a powerful tool that provides insight into the health of your applications around the clock and as they're perceived by your end users worldwide. As HTTP and browser monitors cover the application level of the ISO /OSI model, successful executions of synthetic tests indicate that availability and performance meet the expected thresholds of your entire technological stack. Combined with Dynatrace OneAgent^®, you gain a precise view of the status of your systems at a glance. But is this all you need?

Why browser and HTTP monitors might not be sufficient

In modern IT environments, which are complex and dynamically changing, you often need deeper insights into the Transport or Network layers. Whether necessary as part of deep root-cause analyses of issues faced by your users that impact your business or if you're an engineer responsible for the infrastructure hosting your applications and network paths. You want to be able to answer questions like these:

• What is responsible for application slowdown? Is it a bug in the codebase, a malfunctioning backend service, an overloaded hosting infrastructure, or perhaps a misconfigured network?
• Are all network devices up and running, and is the network providing reliable and swift access to your systems?
• Do I have reliable and up-to-date info on the health of services that are not HTTPS-based but are still crucial elements of your business-critical systems?
• Are the hosts on which OneAgent can't be deployed for technical reasons (for example, a legacy OS) or security-related reasons (for example, systems processing financial information and third party vendors) up and running? Are the corresponding services running on those hosts?

How Dynatrace addresses these questions

Traditionally, Dynatrace provides powerful Platform extensions, which might answer all the above questions. However, those solutions have certain limitations:

• They are not scalable as a pure synthetic solution powered by an API for quickly defining and maintaining tests. Thus, covering an infrastructure the size of several thousand hosts, which is not unusual in modern enterprises, might be significantly challenging.
• Even though these issues were addressed with Extensions Framework 2.0, Synthetic Monitoring still boasts its own advantages with the combination of NAM and HTTP/Browser tests from the same location as well as higher capacity ActiveGates.

To help you enjoy a new and more powerful solution, we're launching a tool that facilitates the automatic transformation of your existing extensions based on EF1.0 into NAM test definitions. Our script, available on GitHub, provides details.

Why Dynatrace NAM is the solution you need

Ease of deployment and management

Synthetic NAM monitors are executed from the same private locations used for monitoring your internal applications with HTTP and browser monitors. This guarantees oversight into high availability, as test executions are load-balanced between nodes hosting private locations.

Also, the API can be used to manage tests at scale, the same way you might use APIs to manage all other synthetic tests. This simplifies solution deployment because NAM is available for current Dynatrace Synthetic Monitoring users and manages tests covering many devices or services.

Test customization

Even though ICMP pings are not the most sophisticated tests, certain customization options can help tailor tests to your needs. For example, you may select the number of packets sent during each execution of the test (to mitigate the risk of temporary anomalies in the network that could impact your test result) and packet size (for example, to ensure that the given MTU is defined on the whole network path).

Compared to other solutions I have tested, Dynatrace NAM monitors are the most configurable which is to my liking. We are already using 3 or more packets within all of my ping tests and see customized packets size as a valuable option for the future.

- Observability Lead, Ivy League University

Of course, it is also possible to define the required success rate of ICMP packets for each test execution against a given IP address to determine if a test is successful.

Similarly, you can adjust your TCP port check tests to validate if selected ports are not opened or closed as intended. Sometimes, a port's desired state is not to accept connections. You might want to use a port to validate if security settings are properly applied and if ports that are supposed to be closed are closed.

Easier test management and integration with infrastructure monitoring tools

A single NAM test can cover one host (for ping) or socket (for TCP port check); however, you can make your life easier by creating single tests covering multiple (up to 1k) hosts at once. Whether you aim to check all addresses with ping tests from a given IP list or range, you can succeed with a single test! Similarly, a single TCP test can check if connections are accepted on multiple ports of one or more hosts

But we went one step further here. Dynatrace NAM stands above all other competitive offerings because you can define a single ICMP ping test covering all hosts within a given host group (optionally narrowing down the IP addresses to a given range). The beauty of this solution is that even if the configuration of the host group changes, the configuration of your synthetic test stays up to date with no extra action needed.

Figure 1. Various approaches to defining targets for NAM tests

Of course, the reporting side is properly adjusted to handle such monitors. For each failed monitor and each problem triggered as a result of such a failure, you can easily identify which requests (IP addresses, sockets) are responsible for the failure and might require appropriate attention.

How NAM tests work

You can use the API or our new web UI (Dynatrace SaaS only) to create NAM tests.

Figure 2. New Synthetic app covering NAM monitors

The new Synthetic app covers the configuration and reporting of network availability monitors thanks to the power of Grail™ and the new Dynatrace platform.

NAM monitors are defined in a similar manner to other synthetic monitors. Apart from elements specific to the given type of test, such as:

• IP address and number of packets for ICMP ping
• Port number for TCP port check
• Names to be resolved and DNS server for DNS tests

All other elements (like assigning locations or conditions for triggering alerts) are the same as those used in other Synthetic monitors.

Figure 3. Example NAM TCP test configuration

The reporting also looks similar; easy-to-understand charts indicate the availability and performance of the whole network availability monitor, and metrics provide an overall summary and links to currently open problems.

For more complex network availability monitors covering multiple requests, you might like to look at the chart presenting the request success rate. This chart shows a list of all requests in the form of a table or a separate chart reporting the executions of separate requests.

Figure 4. Example reporting for a complex TCP test. Overview and detailed requests comparison.

NAM is available for Dynatrace Managed users

While the new, Grail-powered Synthetic app is available only in Dynatrace SaaS environments, Dynatrace Managed users can also benefit from NAM. Dynatrace Managed users get:

• A powerful API for creating and managing synthetic tests (the same as for Dynatrace SaaS users).
• A set of metrics allowing query results with Data Explorer and creating advanced reporting using Dynatrace Dashboards.
• Basic reporting, available out of the box, provides insight into the values of the most important metrics, describing the execution of all monitors and indicating which requests might be causing failures or slowdowns.

"With Synthetic pings functionality we're doing reliable full stack monitoring of network connection between our on premise DC and our infrastructure in cloud."

- Dennis Brouwer, Technical Solution Lead in the Cloud team at Alliander

Figure 5. Example NAM reporting in Dynatrace managed

How Synthetic NAM monitoring affects your billing

NAM is available for all Dynatrace customers. A DPS contract is not a necessary condition for turning on NAM monitoring.

Unlike HTTP or browser monitors, NAM monitors don't have a separate line on the Dynatrace rate card (please contact your Dynatrace account manager for complete details). Instead, you're billed based on the number of metric data points generated during each execution of a NAM test.

The following details apply to metric data points:

• Only the consumption of metrics produced at the request level affects your billing
• Each request execution within ping tests generates 6 metric data points.
• Each request execution within TCP/DNS tests generates 3 metric data points.
• The price stays the same whether you create several tests containing a single request or one test with numerous requests for the same set of hosts or devices.

Next steps

• Get familiar with the new Synthetic app! It fully supports NAM monitors, but a preview of HTTP monitors will start soon.
• Find out how Synthetic NAM monitoring affects your billing on classic and DPS.
• Expect an easy way of integrating NAM monitors with network devices monitored within the IO app (coming soon).
• Configure NAM monitors directly from the Discovery and Coverage app (coming soon).

Public link

Please use the above public link if you want to share this noodl on another website.