What happened ?

At around 12:40, a small update was deployed on our production servers in relation with future feature updates. We received a few notifications from customers stating some agents were unable to log in to their Voice Management consoles.

In the meantime, we got matching internal alerts regarding abnormal failures in the agent connection process.

While connected users didn’t experience any issue, some users trying to connect to their Voice Management console (agents only, not supervisors) may have been unable to do so.

We rolled back the feature by 13:15.

What went wrong ?

While our automated and manual testing performed as expected, we missed a case that was dependant upon users’s call history.

Furthermore, the message displayed to the agents was misleading (login failure, instead of something related to an internal Axialys failure)

What we are doing to prevent this from happening again ?

We are updating our testing process to test a larger set of cases and catch such an issue beforehand.

We will be fixing our login process failure mechanism in order to better reflect the real cause of the issue in the coming days.

We sincerely apologize for the inconvenience.