Insights are a unique feature provided by Kentik. They are autonomous notices that tell you about interesting or anomalous behaviors. Insights can be any fact that is relevant to the traffic at which you are looking. Insights help you quickly recognize where there might be a problem that requires your attention.
This February 2021, we're making Kentik Insights even more powerful.

New Comparison Insights

We are releasing a number of new comparison insights for Cloud, Edge, and Service Provider dimensions. These have been introduced in conjunction with the comparison threshold improvements (explained below). Please also see the Knowledge Base documentation.

Edge Traffic Comparison

Cloud Traffic Comparison

Service Provider Analytics Traffic Comparison

With February 2021 comes another heavy load of new Synthetic Monitoring feature additions: Synthetics Dashboard Widgets, more functionality for Mesh tests, we're now surfacing related Network Flow data within Synthetic test results, as well as many more general point improvements. See for yourself.

Dashboard Widgets and My Kentik Portal Support

Kentik customers now have the option of adding synthetics test results as widgets to their Kentik dashboards and then sharing those widgets with their customers (tenants) via My Kentik Portal. Custom dashboards also let you visualize real traffic charts alongside synthetic test results to help you easily correlate test failures with changes in real traffic patterns.

• Editing any existing dashboard will show users an option to add “Synthetic Test View.”
• Clicking Add will show a modal that allows users to select a synthetic test from the list of tests configured in the account.
• If the selected test happens to be a mesh test, the user has the option of choosing to have the widget either show a mesh or a table of results.

High Density Meshes and Mesh Improvements

The mesh tests are a very popular feature and we continue to respond to the most common UX enhancement requests. We’ve made the following improvements:

• For very large meshes, we’ve added dynamic horizontal and vertical scrollbars when the mesh gets beyond a certain size.
• For “Network Performance Meshes” shown in the Synthetics Performance Dashboard, a label has been added to clearly indicate the 15-minute range. Clicking the link propagates the 15-minute time range so that the test results page shows the same results as the performance dashboard.
• The “Customize” button has been renamed to “Show/Hide” for clarity
• We increased the string length for labels on the left and top edges of the meshes that represented sites to allow the strings to be longer and wrap across multiple lines so as to let users determine which paths/routes are failing with a quick glance of the mesh.
• Hovering over the individual cells of a mesh would show location information as city, state, country, which is not easily actionable when testing between public cloud regions or private sites. This is now enhanced with both pieces of information.

• The Cloud Performance preset meshes are well-received. We improved the UX around parsing their information as well as customization:
  • We added the ability to easily clone them (like the SaaS applications), which allows customers to customize them to make a "yybrid-cloud" mesh out of it for their own specific network performance needs.
  • We also show cloud regions in addition to geo (city/state) information in the labels to the left and the top.

Flow Data in Mesh Test Results

Showing real traffic time-series charts alongside synthetics metrics in the test details is an important feature, but this wasn’t possible to do for site-to-site tests (like agent-to-agent tests or mesh tests) because there was no IP to query based on. We solved this by leveraging the new site-IP configuration that was added to the device configuration. If a site-IP is present it will be used to build the traffic chart and allow the user to jump into Data Explorer to slice and dice the data further.

General Improvements

Several general improvements to the UX have been made in response to customer feedback:

• The workflow to find a failing sub-test’s details (the time-series charts) is simplified by making the agent dots on the map clickable. Clicking them now takes you to the test details:

• Defining and assigning a notification channel in the test creation screen is much more intuitive than it was before:

• It wasn’t clear to users that it takes a while (up to a day) for IP changes to reflect. We added a tooltip for this:

• The point where the day changed on the test timeline was not clear — this is improved visually. Also, the date format shown in the selector now matches the format of the date shown on the timeline:

• The agent version for customer’s private agents is now visible to them. Also, checking a customer’s private agents’ IPs and versions is simplified — the agents’ information is readily available in the table:

• Paused tests now show in historical data.
• Test results legend previously said “ICMP could be blocked.” This has been generalized, as the test could be TCP or UDP.

Synthetics

Kentik’s product and engineering teams continued rapid iteration on Kentik Synthetics. The mesh test display has been greatly enhanced to handle large meshes, and synthetics widgets are now available for dashboards and My Kentik Portal. Flow-based traffic data has also been added to mesh test results. This builds on Kentik’s unique combination of synthetics and real traffic information to help our customers easily choose and set up the right tests.

New Insights

We’ve also added a number of new traffic comparison insights looking for traffic volume changes, changes in volume for cloud services and regions, and changes in OTT dynamics.

Alerting, DDoS & Mitigation

On the DDoS front, we’ve delivered a number of improvements requested by customers. Some are focused on better visibility into mitigation actions and others on usability improvements for policy configuration and alert history.

This January, we introduce the 1st round towards a better way to manage your fleet of kproxy agents.
You can deploy kproxy agents to secure the network telemetry data between your own devices and Kentik's public Flow/SNMP/BGP ingest front-doors, configuring your routing gear to ship this data to a kproxy instance in your network, which will relay the resulting multiplex securely to Kentik's SaaS Clusters.
More information about kproxy here in our knowledge base.

Kproxy is now a first-class citizen in the portal, i.e. it comes with its own Settings screen. This screen comes with the following elements:

• Ability to name and locate your agents in sites that you've configured for other routing devices, and set flow listening IP (mostly useful for config snippets)
• 
• A new onboarding path that will let users deploy a kproxy instance in the first touch experience as a first step in the flow-generating device provisioning first touch onboarding flow.
  
  • The onboarding flow, as well as the modal from the Flow Proxy Agent screen in Settings now displays updated instructions to deploy kproxy both in an ad-hoc manner as well as using the industry-standard systemd service manager.
• The updated deployment method for kproxy now alleviates the need for user-specific API credentials to launch a kproxy instance. Company-global credentials are provided in the instructions, whether in onboarding or on the kproxy settings screen.
• Multiple screens in the existing Kentik UI have been updated to reflect a flow device’s dependency on kproxy as follows:
  • The Network Explorer Device Details page, in the “More Info” section
  • The Settings > Device page with its device details drawer
    
    

Note that existing kproxy-based users will see their kproxies show up in the UI without the site, name and IP metadata associated with them. These can freely be updated as suggested by the UI.

Interface Classification is one of the pillars behind an optimally efficient use of Kentik for Network Observability: it helps our underlying query engine understand what enters, leaves, goes through your network. Via Connectivity Type classifications, it brings business color into your flow data. While programmatic interface classification based on SNMP data is achieved by our Interface Classification engine, we realize that not all of your interfaces always come with consistent descriptions. To solve this problem we're introducing this January a whole lot of additional Interface Management capabilities to help you take classification to 100%.

---

Interface IP Address Overrides

We are now supporting the capability of overriding an interface’s IP address. This override works in the same way the other interface overrides work. Here are some example scenarios:

• The ability to set IP addresses for platforms that do not support obtaining the IP addresses of interfaces from SNMP
• The ability to remove the IP address that is obtained from SNMP
• The ability to set the IP address regardless of what SNMP provides

Static Interface Classification

Users are now able to statically classify interfaces from multiple areas in the product. This feature provides a good way for companies with loosely-enforced interface naming conventions to be able to correctly classify their traffic. This feature allows users to benefit from the full breadth of Network Explorer and all modules relying on Interface Classification.

Important: Interface Classification attributes (Network Boundary, Connectivity Type, Provider) defined statically per-interface will always supersede dynamic rules defined in the Interface Classification engine. As a result, the dynamic interface classification UI will also show when such static definitions are set.

The rules evaluation UI has also changed so that statically classified interfaces appear as such when they escape dynamic rules, as they always supersede them. An orange checkmark will indicate that an interface is statically classified, as displayed in the screenshot below:

From the Settings > Manage Interfaces Screen

Multiple interfaces can be selected and bulk-configured from the top-right Classify dropdown, and users can also access a more complete Interface Attributes config screen from the edit icon button at the end of each interface’s row.

The interface screen’s right-side filter section will assist the user in identifying unclassified interfaces, statically classified interfaces making the manual classification process easier.

Additionally, this panel will also assist users in identifying interfaces that have been statically classified, offering the selectors displayed in this screenshot, prefixed with the “Overridden” mention.

Lastly, users will also be able to reset an interface’s static configuration from the single interface configuration screen, as depicted below, by hitting the “Restore” link:

From Any Network Explorer > Interface Detail Screen

As shown in the screenshot below:

If you'd like to learn more about Interface Classification in Kentik, the best place to start is this knowledge base article.

In lieu of using Kentik’s on-the-fly reverse DNS, customers can leverage kproxy and have it query their own DNS and insert rDNS values into the Kflow data. Among other things, it allows resolution of private IPs reverse DNS, but the initial tradeoff was that it cost customers two of their own custom dimensions. Upon rollout of this feature, native Kentik dimensions will be used.

How can I use this feature?
The following Knowledge Base Article will help you configure kproxy to perform this task.

My Kentik Portal (MKP) is a multi-tenant white-label network observability service that enables Kentik’s customers to market network analytics on top of their existing services. MKP is flexible and easy to use. MKP analytics data can be customized to serve customers’ unique requirements.

What’s New with MKP?

My Kentik Portal has been ported to v4. In addition to everything available in v3, the new v4-based MKP comes with a new and streamlined landlord and tenant UI. Also, v3 was only available to Kentik Premier customers. With v4, we are now making MKP available to all customers (Classic, Pro, Premier).

Read Greg Villain's full announcement in this blog post.

A few noteworthy improvements brought with the v4 version include:

Introduction of Packages & Brandable Tenant Templates

While in practice the templates work the same, landlords now have the ability to brand them. The “Packages” tab will show the adoption of these packages across tenants.

An Overhauled View Assignment UX

Landlords can now see all the views assigned to a tenant at configuration time, whether assigned ad-hoc or as part of a template.

Options for Simpler and Safer Tenant Configuration

Tenant configuration now includes the ability to view the resulting filter as well as a direct option on all dashboards to “Preview as a Tenant.” This will append a clearly identifiable (named) filter group to the current dashboard filtering options and mimic what the target tenant will see.

Redesigned Tenant Portal UX

The redesigned tenant portal takes advantage of the screen’s real estate to display all of the analytics and alerting information at a glance. The improvements also allow for Guided Dashboard parameters to be entered directly from the landing page.