A persistent problem that prevents some customers from accessing their Exchange Online mailboxes through Outlook online is the subject of an investigation by Microsoft.

Despite Microsoft’s claim that the issue only affects North Americans, user reports indicate that it may also affect South American users (Brazil, Chile, the Dominican Republic, and Colombia).
Additionally, some users in South America claim that the Outlook desktop application crashes upon launch.
In a new incident report that has been added to the admin center, Microsoft stated, “Impact is specific to some users in North America who are attempting to access their Exchange Online mailbox using Outlook on the web.”
When attempting to access Outlook online, affected users may receive an unexpected 500 error. We’re examining any new arrangements that could be adding to effect and keep on investigating administration checking telemetry to find the underlying driver of effect.”

Customers reported that they were unable to access the Service Health Dashboard in the admin center under EX610644, despite Microsoft’s recommendation to do so for additional information regarding the ongoing outage.
In addition, Microsoft acknowledged earlier this month that DDoS attacks claimed by a threat actor known as Anonymous Sudan resulted in the suspension of some of its services, such as the Azure Portal, Outlook.com, and OneDrive.
The Outlook.com outage began on Monday evening (June 5) and was resolved by the company in the early hours of Wednesday.
“We know about these cases and are researching. Microsoft told BleepingComputer at the time, “We are taking the necessary steps to protect customers and ensure the stability of our services.”
Updated at 10:30 EDT on June 27: Microsoft has confirmed that the connectivity issues are more widespread than first thought.
Update June 27, 14:13 EDT: After an initial fix failed to restore the service as expected, Microsoft is testing a new one.
Updated at 16:29 EDT on June 27: Microsoft claims that Outlook on the web is back online for all affected users seven hours after the outage began.