Last Thursday, we made the exciting announcement about our plans to join forces with Cisco. Our goal has always been to empower customers to see the Internet and cloud like it’s their own environment. Together with Cisco, we are excited about the possibilities to utilize our intelligence to find new ways to turn visibility and control into action.
Over the past couple of months, our product team launched many new features to provide our customers with the visibility necessary to deliver great digital experiences. Here are some of the highlights:
Internet and WAN Monitoring
One clear lesson from the shelter in place is that you cannot take for granted your reliance on Internet and WAN connectivity and performance. Recently, we delivered a number of important enhancements to deepen the visibility and control across these networks.
- Test Interface Selection for Enterprise Agents enables a network administrator to configure a single Enterprise Agent to monitor the same target via multiple paths. A common use case is monitoring a multi-homed site or enterprise. Multi-homing allows companies to load balance network traffic across multiple paths to reach a destination. However, if a particular path is affected due to a network outage or brown-out, it can be difficult to pinpoint the root cause. Selecting an interface allows our customers to send traffic through each path to the same destination so network admins can quickly identify and remediate the path causing the issue. This is increasingly more prevalent as enterprises adopt internet-centric WANs and use software-defined WAN vendors to facilitate this WAN transformation. Take a look at the example below:
Users in a branch office would often report periodic slowness while accessing a SaaS application. Upon examining per WAN path data to the same SaaS application, it became clear that users would most often experience slowness when traversing a specific ISP (True Internet, in the picture above) used as an underlay for their SD-WAN. Without this data, it would be incredibly difficult to pinpoint this issue and interface selection now provides an easy way to measure each path.
- TLS Certificate Expiration and Session Information was released in late April. HTTP tests now provide session information such as certificate chain, TLS version and cipher suite. This allows customers to configure alerts for the certificate expiration date, SSL version used, or when a weak cipher suite is negotiated.
With the number of very public application outages traced back to improperly maintaining SSL certificates, this information can be invaluable. Sumana from our product marketing team wrote a great blog on this new capability that I recommend reading.
- Path Visualization is at the core of how customers leverage ThousandEyes. We continue to innovate in this area to improve the experience. We also introduced a new path visualization probing mode called ‘in-session’ that enhances how path information is captured when firewalls or other middleware boxes are present in the delivery path. The new path trace mode first initiates a TCP session with the target server and sends path trace packets within the TCP session. You can find this gem under 'Advanced Settings' for your tests.
We also made changes to improve the usability of path visualization by collapsing white nodes (star nodes, where ICMP is blocked by a Firewall) when there are 3 or more in a row. Small but impactful improvement.
- Expanded Enterprise Agent support for Raspberry Pi gives customers additional flexibility to deploy vantage points on a further growing list of supported deployment options. This newly supported mode adds to the existing list of supported types that include physical, virtual appliance, docker, Linux packages, and all major hypervisors. You can read more about supported Enterprise agents here.
Over the past few months, we’ve released a number of significant enhancements to help customers improve their visibility and correlate app and network performances that can impact the user experience. More and more modern web apps comprise a large number of 3rd party services and API calls. For many apps, the sequence of API calls between 3rd party services across the Internet can have a large impact on the ultimate user experience.
Building upon the new synthetic transaction engine that we introduced last October, we recently added support to monitor and interact with arbitrary API endpoints from within a transaction script. For example, making HTTP requests and initiating TCP and TLS connections directly from the transaction scripts’ execution environment. This new capability for transaction tests can be used in a number of new ways, including:
- Surrounding a fetch HTTP request with markers to measure the responsiveness of an API endpoint.
- Testing arbitrary and proprietary network protocols.
- Making POST calls directly to an API endpoint.
- Mixing API calls and browser commands to, for example, create a resource via API and then view its data through the web application's frontend.
- Using an API endpoint to generate an ephemeral access token for use in a transaction test.
End-User Experience Monitoring
Recent polls put the percentage of people working from home above 60%. During these challenging times, employees are doing their best to continue to get their jobs done, where they are connecting to business-critical apps across VPNs or to a growing number of SaaS applications. This new normal creates unprecedented challenges on IT and Workplace teams to get adequate visibility to help support employee’s productivity. What used to be a finite set of predictable and known environments (offices, corporate wifi networks, ISPs, etc) to manage has now exploded. Every user’s end to end connection path is a snowflake and troubleshooting a user’s machine is no longer as simple as walking over to their desk. To support customers in March this year, we launched our remote user experience offer that provides use of our Endpoint Agents free, for 90 days. During this time we have also introduced a number of important enhancements:
- Improved bulk management of Endpoint Agents in-product and through the API. We have improved the administrative experience when navigating across thousands of endpoint agents. We have also introduced new APIs to query the last seen IP address and modify the endpoint agent name.
- Enhanced forwarding loss for Endpoint Agents in path trace views. This is one of the key metrics to help identify a node in the network path that might be affecting the user experience. This improvement enhances an admin’s ability to effectively localize an issue across a user’s end to end experience.
- Collect browser session metrics from Microsoft Edge is now available. Browser extensions are used to capture real user measurements from browser sessions and provide correlation to the underlying network conditions during the browser session. This new support for Microsoft Edge (version 79+) sits alongside our existing browser extensions available for Google Chrome and Microsoft Internet Explorer 11.
If you have been following our Internet Report podcast or the Internet Outages Map, you will have seen first hand how unpredictable the Internet can be. Since first introducing Internet Insights™ in November of last year, we have been working hard to continuously expand the provider catalog and deliver authoritative intelligence for macro-level Internet outages. Several key Internet Insights enhancements have been added:
- New event metrics and breakdown for outages to improve insight into the impact of outages especially when they are long or occurring at the same time. Outage start, end, and duration metrics have been added to table and topology views. Outages occurring at the same time are now presented as a list when hovering on the timeline or topology. A purple outage swimlane has been added to the timeline to improve the visibility of long and overlapping outages.
- Three new agent grouping options to improve the way agents are displayed and organized in the Topology view: Origin Network (groups agents by their origin network), Next Hop Network (groups agents by the network in the next hop), and Agent (lists agents sorted by the number of affected tests).
- New destination grouping option to improve the way destination interfaces are displayed and organized. A new IP Prefix grouping option has been added. This allows users to filter the destination view so that only nodes with IP prefixes are shown.
- Outage alerts included in the alerts API to help ensure you are notified in real-time when outages occur. See our alerts API documentation.
- One-click from test snapshots to Internet Insights. Test snapshots affected by an outage now let logged-in Internet Insights users click through to the outage event impacting the test.
- Expanded provider catalog with additions and updates across IaaS, ISP, CDN, DNS, UCaaS, and SECaaS provider types. See the changelog for details.
To help customers improve the way they aggregate and act on the data collected from ThousandEyes, our platform teams have recently introduced a number of enhancements.
- Dynamic baselines help customers ensure the alerts they receive are actionable for a wide range of scenarios. Dynamic baselines allow users to create alerts that more accurately reflect the natural variance in test data. Using standard deviation, percentage change, or absolute values, users can configure alerts that dynamically determine whether to fire or not, based on historical data within a sliding time window.
- Sticky agent alert rules allow users to configure alerts that only trigger when the same agent/s meet a specific condition in any number of consecutive rounds. This helps to alert in a scenario where only one agent is experiencing an issue.
- Expanded Dashboard APIs and support for Color Grid widgets for reports are now available. The new API allows full scripted automation to simplify the task of creating, updating, and deleting dashboards. Very useful when a large number of dashboards are required or when looking to replicate dashboards across account groups.
Thank you to all of our customers for their continued insights on how we can improve ThousandEyes. Our promise is to continue to develop our products with a relentless focus on solving hard customer problems and innovating without boundaries.