Classic Outlook on Windows: Automatic Replies and free/busy fail on local PCs but work on the terminal server

User A:P 0 Reputation points
2026-11-19T12:17:41.83+00:00

Hi everyone,

I’m troubleshooting an issue that initially affected one user and now affects a second. Both use classic Outlook for Windows with Exchange Online.

On the affected PCs, opening Automatic Replies gives this error: “Your automatic reply settings cannot be displayed because the server is currently unavailable.” When either user creates a meeting, Scheduling Assistant also fails to show colleagues’ free/busy information.

Both features work for the same users in classic Outlook on our terminal server and in Outlook on the web. I also tested my own account on one of the affected PCs and reproduced the issue there.

So far, I’ve repaired and reinstalled Office, created a new Outlook profile, and tested with a new Windows profile. The issue persists. I also compared Outlook’s Test E-mail AutoConfiguration results on the affected PC and the working terminal server. The Autodiscover results looked the same.

I found an older thread where changing the organization’s EWS application access policy resolved the same symptoms. In that case, everyone was affected. In ours, the same mailbox works from the terminal server.

Has anyone seen this pattern? Which Outlook log or diagnostic would you use to compare a failed Automatic Replies request on the local PC with a successful one on the terminal server? I’d like to see where the request fails before considering any organization-wide change.

Thanks for your help! User's image

Outlook | Windows | Classic Outlook for Windows | For business
0 comments No comments

2 answers

Sort by: Oldest
  1. AI answer

    2026-11-19T12:18:11.2+00:00

    The most relevant diagnostic from the available guidance is to compare the Autodiscover/EWS path between the affected PC and the working terminal server, then check the server-side logs that capture free/busy requests.

    For free/busy, use this comparison path:

    1. On the affected classic Outlook profile, run Test E-mail AutoConfiguration.
      • Hold Ctrl, right-click the Outlook icon in the notification area, and select Test E-mail AutoConfiguration.
      • Test Autodiscover for the affected user and note the returned EWS URL.
    2. Run the same test on the working terminal server session for the same mailbox.
      • Since Autodiscover already looked the same in the comparison, that points away from a simple Autodiscover mismatch.
    3. If free/busy is failing, validate the EWS endpoint directly.
      • Use the EWS URL returned from Autodiscover.
      • Browse to that URL or run Invoke-WebRequest -Uri "<endpoint URL>" -Verbose.
      • Preferably test from an external network when validating the published URL.
    4. If the issue is in a hybrid or published path, compare the request handling in logs.
      • Check IIS logs for free/busy requests that return 503 Service Unavailable in %SystemDrive%\inetpub\logs\LogFiles\W3SVC1.
      • If the request is not in W3SVC1, check %SystemRoot%\System32\LogFiles\HTTPERR.
      • These logs help identify which server handled the request and whether IIS or another web component returned the failure.
    5. If a load balancer is in the path, compare requests through it with a controlled test to the correct backend.
      • Keep the published hostname and certificate validation.
      • Do not leave temporary routing or Hosts-file overrides in place after testing.

    For Automatic Replies, the available guidance confirms a known classic Outlook issue where opening Automatic Replies can fail with:

    “Your automatic reply settings cannot be displayed because the server is currently unavailable. Try again later.”

    That issue was first observed with Current Channel Version 2505 (Build 18827.20128), occurs starting with builds 18510.10000+, and is fixed in Version 2508 Build 19127.20000.

    Given that the same mailbox works in Outlook on the web and on the terminal server, the first comparison to make is the Office channel/build on the affected PCs versus the terminal server. If the PCs are on an affected build and the terminal server is not, that directly matches the known issue pattern for Automatic Replies.

    Supported workaround for Automatic Replies:

    • Set automatic replies in Outlook Web Access by opening the shared mailbox in a separate browser window, then Settings > Account > Automatic replies.
    • For classic Outlook, switching Office to Semi-Annual Enterprise Channel avoids the bug until the fixed build is installed. After changing channel, use File > Office Account > Update Options > Update Now.

    Based on the documented troubleshooting, the best logs/diagnostics to compare are:

    • Test E-mail AutoConfiguration output, especially the returned EWS URL
    • IIS W3SVC1 logs for free/busy requests
    • HTTPERR logs if the request never reaches normal IIS logging

    The context does not provide a separate Outlook client log specifically for comparing failed versus successful Automatic Replies requests on two machines. The strongest documented path is to compare build/channel for the Automatic Replies symptom and compare Autodiscover/EWS plus IIS/HTTPERR logging for the free/busy symptom before making any organization-wide EWS policy change.


    AI-generated content may be incorrect. Read our transparency notes for more information.

    Was this answer helpful?

  2. Michelle Nguyen 3,125 Reputation points Independent Advisor
    2026-11-19T13:48:02.37+00:00

    Hi @User A:P

    I'm just a fellow community user, so I don't have access to your organization's environment. Please treat the following as general suggestions.

    You've narrowed this down very thoroughly. As far as I know, in classic Outlook both Automatic Replies and free/busy information go through Exchange Web Services (EWS), which is separate from the regular path for sending and receiving email. Since the same accounts work on the terminal server and in Outlook on the web, and your own account also fails on an affected PC, I think the cause is more likely on these PCs or their network path (for example a proxy, firewall, SSL inspection tool or TLS configuration) than on the mailboxes or your organization's EWS policy. However, I can't confirm this yet.

    To compare a failed request with a successful one, you could try:

    1. Capture HTTPS traffic with Fiddler while opening Automatic Replies on both machines, to see which error code the EWS request returns, and whether it's blocked by a proxy or fails because of a certificate error.
    2. Open the EWS URL (from the Test E-mail AutoConfiguration results) in a browser on an affected PC, to check whether it connects and shows a sign-in prompt.

    To help narrow down the cause, could you please share:

    • Do the affected PCs and the terminal server go through the same proxy or network path? Do these PCs have a different proxy configuration (including WinHTTP proxy) or PAC file?
    • Is there any security software, VPN or SSL inspection tool installed only on the affected PCs?
    • Are the Outlook version and update channel the same on the affected PCs and the terminal server (File > Office Account)?
    • When did the issue start, and did it coincide with any Windows or Office update or network change?
    • Are any other PCs on the same network affected in the same way?

    I agree with your approach of checking the PC side thoroughly before changing the EWS policy for the whole organization. Please update the information above.

    Was this answer helpful?


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.