Response to Coronavirus
The COVID-19 virus is putting a huge strain on IT and network teams all around the world. To help with the increased surge of remote workers, we at ThousandEyes are offering our Endpoint Agent product, free of charge. For more information, please see our blog or contact the ThousandEyes team.
Migration of Transaction (Classic) Tests
In the next release, ThousandEyes will be disabling the ability to create additional transaction (Classic) tests from the product.
ThousandEyes will be working with individual customers as well as making an automated conversion process from within the product console available to assist with the migration process. This tool is expected to be available in May. If you need the ability to create additional transaction (Classic) tests after April 30, contact the ThousandEyes support team.
Amazon Linux 2 Support
ThousandEyes now supports installing Enterprise Agents on Amazon Linux 2 operating systems.
New Documentation Platform
ThousandEyes is excited to announce that at the end of April, we will be launching a new website for product documentation. This is the first stage in our efforts to provide users with an improved content experience.
New Features and Enhancements
The new “sticky agent” feature allows a user to configure alerts that only trigger when the same agent/s meet a specific condition in any number of consecutive rounds.
For example, an alert rule is configured for if the same agent trips a specified threshold in three consecutive rounds. The Atlanta cloud agent trips the rule in round one, the Ashburn cloud agent trips it in round two, and the San Francisco cloud agent trips it in rule three.
In this scenario, the alert rule would not trigger when using sticky agents. Either Atlanta, Ashburn, or San Francisco would need to trip the rule in three consecutive rounds to trigger the alert.
This can be used to configure alerts about the health of the network in a specific agent’s location, to increase confidence in the accuracy of alerts being sent (for example, issues that are seen from the same set of agents across multiple rounds), or to improve the signal to noise ratio by reporting on agent health as opposed to only test health.
For more information, see How Alerts Work.
Cloud and Enterprise Agents
- Improved the warning message that notifies users when an agent set is mixed, as bandwidth measurements are not allowed to run from Cloud Agents.
- The Endpoint agent now supports Microsoft Edge version 79 or higher for collecting browser session metrics. To install the extension:
- On the desired endpoint, launch Microsoft Edge.
- Open the following link: https://app.thousandeyes.com/install/endpoint-agent
- The createdTime, lastSeen and version fields are now exposed in the /endpoint-agents API.
Several improvements have been made to how Internet Insights displays outages. Outage event metrics (begin, end, and duration) have been added in the following areas of the UI:
A purple outage swimlane has been added below the timeline to improve visibility of long outages and outages that cross a five-minute period.
A tooltip appears when hovering over events on the swimlane, improving discoverability of individual outages and of when multiple outages occur at the same time.
Views > Topology
In the text displayed while hovering over an affected interface node, a list of outages is now displayed when multiple outages occur at the same time.
Views > Table:
Outage event metrics have been added to the Table view.
- In some instances, an issue occurred when a transaction script failed for a reason other than a test timeout that resulted in the error not being reported correctly. This has now been resolved, and detailed error messages should be seen for all failed transaction scripts.
- Previously, there was no option in Transaction test settings to enable following redirects in the HTTP server layer of the test. This has been corrected.
- Fixed an issue that only allowed Latin1 characters in the post body; all UTF-8 characters are now allowed
- Fuzzy searching in the agent selector in test settings has been disabled due to unexpected results.
- DNS Server snapshots that had a server selected were producing an incorrect "Invalid URL parameters" popup. This has been corrected.
- The link to the "Getting Started" videos in the Help & Support dropdown was not loading correctly. This has been fixed.
- Previously, creating a test with a duplicate test name failed with no feedback. The intended error message has been restored.
- An issue occurred when highlighted controls in the Device Topology view were not updating when the selected metric was changed. This has been fixed.
- An issue occurred where organizations with large numbers of Endpoint Agents could not use labels to correctly filter their data in dashboards and reports. This has been resolved.
- Previously, the AlertList widget included DNS+ alerts, but the AlertGrid widget does not. This lead to inconsistencies when looking between the two widgets. DNS+ alerts are now available in the AlertGrid widget.
- Endpoint Agent links inside the map widget were incorrectly pointing to the Enterprise Agent list. This has been resolved.
- An issue occurred when too much data resulted in the browser being unable to render the Path Visualization. This following corrections have been made to resolve this issue:
- The error message displayed in the UI has been corrected, and now states “The path trace graph is too big, try adding some filters.”
- The default grouping has been set to “Agents” in all Endpoint Agent view scenarios.
- An issue occurred where Path visualization controllers ( Filter, Group By, and Number of hops etc) were not displayed when there was too much data. This has been resolved.
- An issue occurred in the network topology where trying to apply any filter would result in a spinning wheel and not return any options. This is now fixed.
- An issue that caused the agent label color to revert to gray when the name was edited has been corrected.
- Previously, the Verify Content checkbox stayed enabled after removing specified content when setting up a new test. This is now fixed.
- An issue occurred with some path visualization filters where search options were filtered out and not displayed as they were not on the current pagination page. This issue has been fixed.
- An issue when creating / modifying a test resulted in the custom user agent string disappearing from the UI after saving. This has been fixed.
- The support chat window was not visible on the Endpoint Agents → Overview page. This has been corrected.
- An issue occurred with the Internet Explorer extension where hovering over an element in the browser session waterfall displayed the HTTP method as UNKNOWN. An error message to show that headers for the IE extension are not supported.
- A recent change to the user profile created during registration allowed non UTF-8 characters to be used in a username, causing the username to fail to register. This has been corrected, and only valid UTF-8 characters can now be submitted.
- An error occurred when opening some share links where an incorrect popup error would occur. This has been fixed.
- An issue occurred where clicking the gear icon next to a scheduled test did not filter and expand the scheduled test settings panel. This has been resolved.
- Fixed an issue where Endpoint Agents with connection failures were not showing a severity color.
Questions and Comments
Got feedback for us? Send us an email!