MS Teams Conferencing Failures

We have seen that our error rate provisioning MS Teams Conferencing has returned to normal, low levels since 16:10 UTC yesterday. Normal levels of service have resumed, and we're working to improve our monitoring so that similar issues are detected sooner in the future.

We believe the root cause is a lack of generated URLs for a meeting when we call Microsoft Graph APIs. We are waiting for a response from Microsoft on the issue. Unfortunately the timing of this is out of our hands.
We are also working on alternative ways to provision Teams meetings, but this will not provide a short term fix.

We are continuing to work on finding a way to resolve, mitigate or provide an more reliable path to Teams provisioning. We will update this incident over the coming days as we make meaningful progress.

We've found some evidence that other users of the Microsoft Graph API are experiencing the same issue; we're looking for anything that may let us mitigate the impact and we're reaching out to Microsoft for support.

We are still investigating; we are capturing additional logs to help us narrow down the failing population. Some conferences are being provisioned after automatically retrying, but still not as many as we'd expect.

We are continuing to investigate - the issue is not impacting all customers but we are not yet seeing a pattern to the domains or events impacted.

We are investigating an elevated level of failures for provisioning MS Teams Conferencing.