A Microsoft outage that hit Outlook, Teams and Copilot on Thursday was still disrupting services on Friday, affecting government agencies and businesses from Australia and Japan to the United States.
For users whose emails never arrived, the answer in most cases is that they have not been lost. Email sent during an outage is normally held and retried rather than discarded, so it tends to arrive late instead of disappearing, Nacionale News reported.
How email delivery works during an outage
When a sending server cannot hand over a message, the internet’s email standard tells it to keep the message and try again later. That standard, RFC 5321, says servers should generally wait at least 30 minutes between attempts and keep trying for at least four to five days before giving up. Each email provider sets its own schedule, so how long a message is retried depends on who sent it. In practice, mail that could not get through should arrive in bursts as service recovers, sometimes hours late and out of order.
Emails users sent
If Outlook could not reach Microsoft’s servers, a message most likely never left the user’s device. In classic Outlook, an unsent message waits in the Outbox. According to Microsoft’s support page, once reconnected, users may need to open the message and send it again, or choose Send/Receive. In the new Outlook, they can force a manual sync with the Sync button on the View tab.
Once service is back, users should check the Outbox, Drafts and Sent Items. A message in Sent Items has left the account. One still in the Outbox or Drafts has not. Checking Sent Items first also avoids sending the same email twice.
Emails Microsoft accepted but couldn’t deliver
Some messages reached Microsoft’s servers but stalled on the next step. Microsoft’s documentation says Exchange Online, its cloud email platform, keeps trying to deliver such messages for 24 hours before returning them to the sender as a bounce, known as a non-delivery report.
On Friday, Microsoft said it was extending its email timeout window to 48 hours “to help protect delayed messages while service recovery continues”, CNN reported.
If a message does run out of time, the bounce carries Microsoft’s error code 550 4.4.7 and the line “QUEUE.Expired; message expired”. The bounce itself tells users to contact the help desk if the problem continues.
Emails sent to users
The sender’s server will keep retrying. If it eventually gives up, the bounce goes to the sender, not the recipient.
What to do now
- Avoid repeatedly pressing send while Outlook is failing. Wait until it connects, then check Sent Items before sending again.
- For anything urgent, use another channel, such as a phone call, a text message or a different email account, until service is back.
- If a bounce says a message has expired, send it again once Outlook is working normally.
- If something important has not arrived once service returns, ask the sender whether they received a bounce.
- IT administrators can follow the incident in the Microsoft 365 admin center, where a queued messages report shows outbound messages held for more than an hour.
What Microsoft has said
Microsoft has confirmed that email delivery is delayed, but it has not said whether any messages were lost, nor how many were affected. It also has not said whether incoming mail from outside senders was turned away or simply held.
The company blamed server problems and said it was not aware of any malicious activity. Its status updates pointed to a large number of servers becoming unexpectedly unavailable after failing automated health checks, an overwhelmed process that sets up secure connections, and a recent network configuration change that it began reversing. It has not published a root cause.
Microsoft said features of its 365 services that depend on Exchange Online were affected, while those that do not rely on it continued to work.
When will services be back?
Microsoft has not given a time for full recovery. On Friday, it said it was seeing early signs of stabilisation after rolling out fixes more broadly, although service capacity had not returned to normal.
In its latest updates, the company said traffic was being gradually restored. It had completed recovery work across parts of the affected infrastructure in the Asia Pacific region and was now focusing on extending it to other regions. It has not declared the outage resolved.
(With information from Reuters, Nacionale, CNN)
