Our Product and Engineering teams cranked out a slew of new features in Synthetics, just in time for the holidays, so we’re happy to kick the year off with an exciting product update. And given the blistering pace of updates our Synthetics product has been getting, we figured it could use its own dedicated update — so here it is!
There’s a tab for that!
Do you have applications deployed in multiple availability zones of a given cloud provider? How about across different cloud providers? We heard our customers say things like, “I want to know if users on the east coast are seeing problems loading my app because the latency to us-east-1 has increased significantly” or “How can I be sure that my app instance in EC2 is able to reach my BigQuery instance in GCP?” and we realized that not only can we provide a real quick way of answering that question, but we could go a step further and make that answer available to all our customers for free (exactly the way we did with the Free SaaS application performance tab). Say Hello to the new Cloud Performance Mesh tab!
The new Cloud Performance tab gives users a general sense of cloud performance without them having to create these test themselves and needing to use test credits.
We have four preconfigured tests — one each for testing performance between different availability zones of the three biggest cloud providers (AWS, Azure and GCP) and one for testing performance between them.
We hope you find these cloud performance meshes useful. Based on the feedback we receive, we will consider adding similar (preconfigured/ free) meshes for other cloud providers.
Kentik Synthetics users love the path view, and they tell us that it helps them find issues in their network that they were blind to before. Given the high adoption of this feature, we are especially attentive to user feedback. One of the things we heard was, “I love being able to see all the nodes and links, but sometimes I just want to see a list of all the IPs, good old traceroute style!” And so we made that happen!
Clicking on any agent in the path view will pop out a pane on the right (you may recall this new pane from our December update) that will show you the text trace output from that agent to each destination in individual text boxes.
Another point of feedback on the path view was that, while it is very rich in data and in its ability to crunch through that data, the view can get very busy and hard to parse when there are a lot of nodes, and it can become like trying to find a “needle in a haystack.”
To help “de-noise” things, we had (and still have) a way to “collapse all ASNs,” but that was an “all or nothing.” So, now we have improved on that — we are offering customers a way to peel off node layers selectively from either side. Unlike competitive solutions, we can collapse each layer of nodes down to the ASNs and highlight any problems in an aggregated fashion. So, if you collapse a set of nodes and one of them happens to have high loss, you’ll still retain the information about it by looking at the color of the ASN that contains it (turning red).
To increase RBAC for Synthetics, we have built per-user overrides, putting control in the super-administrators’ hands. Super-administrators can selectively restrict test viewing privileges to administrators. We also added similar options for being able only to view but not add or change agents.
Kentik Synthetics now supports being able to tag both agents as well as tests with Labels. A global set of labels can be defined and used across all agents and tests, and existing or new (flow) device labels can also be applied. This feature is currently in limited availability, so if you are interested in checking it out, please reach out via your account team.
Users can now define any number of custom labels and tag them to agents (both private and global/ public). Doing so makes it easy to select groups of agents when creating tests, thereby reducing the time needed to set up tests in addition to reducing errors and maintaining consistency of testing.
Tests can also be tagged with labels, similar to the way agents can. Once tagged, it is easy to organize and sort through the list of tests using the labels filter.