We have identified the issue to be a large influx of Graph API traffic received from Microsoft which has caused our Microsoft Teams recording servers to temporarily overload and not have any capacity to capture and process any additional data. This impact lasted just 4 seconds.
We have been monitoring the impacted cluster for a few hours and we have not noticed a recurrence of the same issue. The platform is back to being fully operational.
Some customers may have noticed that their Microsoft Teams recorded users may have had their conversations forcibly terminated or not allowed to continue, if a fail-close Microsoft Teams type policy was assigned, and some customers may notice a loss of recording within the impacted time window if a Microsoft Teams fail-open type policy is assigned.
If you require further information or a full RCA for this incident, please raise a case with Luware Support by going to our support portal - https://helpdesk.luware.cloud/
We sincerely apologize for any inconvenience caused.