A significant outage affecting Microsoft Exchange Online disrupted access for Outlook on the web users. The incident, which began on March 19, 2025, caused widespread service disruption for a considerable period. Thousands of users reported issues on DownDetector, indicating problems logging in, accessing the website, and establishing server connections.
Microsoft acknowledged the problem, stating, “We’re investigating a potential issue with Exchange Online and checking for impact to your organization,” adding that the disruption affected “any user attempting to access Outlook on the web.”
The company identified a code error as the root cause, explaining,
“We’ve identified a portion of code which is throwing an error. We’re working to revert the recent code change and reviewing data from the reproduction to determine any additional next steps.”
Affected users encountered a “Something went wrong” error message. The outage was tracked as a critical service issue (EX1036356) on the Microsoft 365 Admin Center.
A separate, related incident (EX1035922), also stemming from a recent code update, prevented some users from performing searches within Outlook on the web or the new Outlook client. Microsoft described the error message as “‘We didn’t find anything, try a different keyword’,” suggesting a workaround using filters in search criteria.
Microsoft also addressed a lingering issue from a week-long Exchange Online outage (initially EX1027675, later EX1030895). This ongoing problem caused delays and failures in sending and receiving emails, particularly impacting “a small subset of messages” resulting in NDR failures and intermittent plain text calendar invites with winmail.dat attachments.
By the afternoon of March 19th, Microsoft reported improvements after reverting the problematic code change.
A final statement confirmed the resolution:
“A recent configuration change intended to improve the use of browser content policies within Outlook on the web, introduced a code regression resulting in impact,” the company explained.
“We’ve successfully reverted the problematic code change and verified the issue has been remediated through customer confirmation and service health telemetry.”
The company linked this series of outages to code issues, following a weekend Microsoft 365 outage impacting Outlook and Exchange Online authentication and a subsequent outage affecting Teams and causing call failures.
Helpful Reads:
