Access problem at Microsoft

According to the data of the Downdetector site, which monitors service outages, user notifications were recorded regarding problems accessing Microsoft 365, Outlook and Teams services as of 20.30 GMT.

In the statement made by Microsoft, it was stated that it was detected that the load on a certain part of the infrastructure increased in more than one region, and this may have caused connection errors.

In the statement, it was stated that all possible ways were evaluated to reduce the impacts, all distribution operations were stopped, additional traffic from the affected infrastructure was directed to other points and the affected servers were restarted to put them back into service.

In the statement, it was reported that the increased load was not due to the increase in traffic, that it was determined that the unexpected shutdown of some servers increased the load on the remaining infrastructure, and that efforts were continuing to safely put the servers back into service.

In the statement, it was emphasized that a configuration change was tested in order to fix the problem that caused the servers to fail automatic health checks and become disabled, and it was stated that the process of restarting all affected servers was started simultaneously and their status was monitored while the servers were put back into service.

In the statement, it was noted that an alternative service health check method has been validated, this method has enabled some of the affected servers to be returned to service, and the change in question has begun to be implemented on a broader scale.