Share via

Global mailbox access blackout (EX1253275), why did MAPI/HTTP fail while some OWA sessions stayed alive?

Henry Harris 50 Reputation points
2026-03-17T11:42:00.4066667+00:00

Our users couldn't access their mailboxes via Outlook desktop or mobile.  It was weird because users already logged into Outlook on the Web (OWA) were mostly fine, but anyone trying a fresh login or using the desktop app just hit a wall of connection timeouts. It felt like the protocol layer that handles desktop connections just stopped talking to the mailbox databases.

Was this strictly a failure in the MAPI-over-HTTP protocol handler,

Exchange Online
Exchange Online

A cloud-based service included in Microsoft 365, delivering scalable messaging and collaboration features with simplified management and automatic updates.

0 comments
Answer accepted by question author
  1. Andy David - MVP 160K Reputation points MVP Volunteer Moderator
    2026-03-20T11:12:23.0366667+00:00

    This was a failure on the ExO side, so nothing you could have done.

    According to the PIR:

    A small number of Exchange Online backend servers experienced intermittent network communication failures within their high availability groups. When these servers were unable to communicate their required health signals, built-in protection mechanisms automatically initiated failovers to healthy servers to prevent a broader outage. While only a limited portion of backend capacity was affected, multiple failovers occurred in close succession, which temporarily reduced overall service quality.

    2 people found this answer helpful.
    0 comments
Answer accepted by question author
  1. Steven-N 22,485 Reputation points Microsoft External Staff Moderator
    2026-03-17T13:12:40.4+00:00

    Hi Henry Harris

    Regarding your concern: Was this strictly a failure in the MAPI-over-HTTP protocol handler?

    My answer is NO, this pattern usually isn’t a pure MAPI‑over‑HTTP handler failure, even though MAPI/HTTP is the protocol that appears broken.

    Based on how Outlook, OWA, and Exchange Online behave, what you observed is a front‑end client connectivity failure that disproportionately impacts MAPI/HTTP, while OWA continues to function because it uses a different access path and session model.

    From my research, a practical way to troubleshoot and explain this scenario is to focus on the MAPI over HTTP connectivity path and its prerequisites:

    1. Verify the MAPI/HTTP path is reachable Ensure your firewall/reverse proxy/load balancer is not blocking or misrouting the MAPI over HTTP virtual directory (/mapi), because Outlook relies on this endpoint.
    2. Fix the common Outlook-side block On affected PCs, check HKEY_CURRENT_USER\Software\Microsoft\Exchange\MapiHttpDisabled. If it’s 1, set it to 0 (or delete it) and restart Outlook.
    3. Confirm MAPI isn’t disabled for the mailbox (Exchange Server) If MAPI was disabled at the mailbox level, re-enable it (for example via Set-CasMailbox <user> -MapiEnabled $True) and retest Outlook.

    Hope my answer will help you.

    If the answer is helpful, please click "Accept Answer" and kindly upvote it. If you have extra questions about this answer, please click "Comment".   

    Note: Please follow the steps in our documentation to enable e-mail notifications if you want to receive the related email notification for this thread.  

    1 person found this answer helpful.

2 additional answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. Q&A Assist
    2026-03-17T11:42:17.3566667+00:00
    AI generated content. This question contains an answer created with AI from Q&A Assist. Learn more

    The outage pattern described matches a failure in the MAPI/HTTP path (and related client connectivity components), not in OWA itself.

    OWA connects to Exchange mailboxes through server-side web components, while rich clients such as Outlook desktop and mobile use MAPI over HTTP (or, historically, RPC over HTTP). When the MAPI/HTTP layer or its dependent app pools or services are unhealthy, Outlook clients can be unable to establish or maintain connections even though OWA continues to function for already-established browser sessions.

    In on-premises scenarios, a known cause of intermittent or failed MAPI/HTTP connectivity is unhandled exceptions in the MAPI-related IIS application pools, such as MSExchangeMapiMailboxAppPool or MSExchangeMapiAddressBookAppPool. When these app pools fail, Outlook clients using MAPI over HTTP are disconnected or cannot start, while other protocols (including OWA) may remain available.

    Additionally, if MAPI/HTTP is disabled or not functioning correctly, Outlook may fall back to legacy RPC over HTTP where possible. However, RPC over HTTP is no longer supported for Exchange Online and is being actively deprecated. For Office versions that support MAPI/HTTP, registry-based attempts to force RPC over HTTP can be ignored, reinforcing that desktop connectivity depends on a healthy MAPI/HTTP stack.

    Therefore, the behavior observed—desktop and fresh client logons failing while existing OWA sessions mostly stayed alive—is consistent with a failure or degradation in the MAPI-over-HTTP protocol handling path (and its underlying app pools/services), rather than a complete mailbox or OWA failure.

    References:

    1 person found this answer helpful.
  2. Deleted

    This answer has been deleted due to a violation of our Code of Conduct. The answer was manually reported or identified through automated detection before action was taken. Please refer to our Code of Conduct for more information.

    Comments have been turned off. Learn more

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.