Why Exchange Online returns 550 4.4.7 QUEUE.Expired; message expired after retrying for 24 hours, how to tell a destination-side problem from your own, and how to check MX and DNS, TLS certificates, connectors, hybrid routing and duplicate recipients.
Home › Guides › Exchange Online 4.4.7
Last updated 17 November 2026
Microsoft documents the diagnostic line of this NDR as:
#550 4.4.7 QUEUE.Expired; message expired ##
with the user-facing text "The server tries to deliver this message but can't. It stops trying. Try sending this message again. If the problem continues, contact your help desk."
The receiving server kept answering with a temporary (4.x.x) error, or couldn't be reached, instead of accepting or permanently rejecting the message. Exchange Online repeatedly tries to deliver the message over 24 hours, and only after 24 hours of unsuccessful attempts does the sender get this NDR. Microsoft lists four causes:
First question: does it happen for one destination domain or for many? One domain points to the recipient's side. Many domains, or your own recipients, point to your configuration.
The diagnostic section usually names the last server tried and the error it returned. Microsoft calls out this pattern:
450 4.7.0 Proxy session setup failed on Frontend with '451 4.4.0 Primary target IP address responded with ...
Record the error that follows that string and the last endpoint attempted. That's the evidence the other side's admin needs. Paste the NDR headers into the Email Header Analyzer to see the hops and any server names.
| What you see | Likely meaning |
|---|---|
| QUEUE.Expired after a remote server name and a 4.x.x reply | The destination server kept deferring the message for 24 hours. |
| Primary target IP address responded with ... (connection or timeout error) | The destination host is down, unreachable or blocking connections from Exchange Online. |
| Transient Failure during recipients lookup AmbiguousRecipientTransientException | Your own directory: Exchange Online found more than one object for the same recipient (see the duplicates section). |
| 450 4.4.317 Cannot connect to remote server [Message=UntrustedRoot] | A TLS certificate chain problem on the remote server (see the TLS section). |
Microsoft lists these destination-side causes: temporary network or internet problems at the destination, an incorrect DNS record (for example a missing or wrong MX), and aggressive anti-spam settings that block legitimate senders, such as all senders from Exchange Online.
When the destination admins find nothing wrong, Microsoft lists these fixes for the sending Exchange Online organization:
If the 4.4.7 NDR says "Transient Failure during recipients lookup AmbiguousRecipientTransientException", Exchange Online couldn't uniquely resolve the recipient and treated the ambiguity as temporary until the message expired. Microsoft lists these causes:
Find the objects that share the address, remove or change the duplicate, and let sync finish before you resend.
A remote server that sends only its leaf certificate, without the intermediate CA certificates, can break delivery. Microsoft documents NDR 450 4.4.317 Cannot connect to remote server [Message=UntrustedRoot] for this: Exchange Online doesn't fetch intermediate certificates on demand, so a remote MTA must present the full chain. The remote admin fixes it by configuring the full certificate chain. If you are the Exchange Online side, tell them about the NDR.
Also check any partner connectors for the destination domain (Exchange admin center, Mail flow > Connectors). A connector that requires TLS, or routes to a smart host, only works if the other side really supports that: Microsoft says TLS-required partner flows need a valid certificate from a commercial CA trusted by both parties. A partner who changed servers, IPs or certificates can leave your connector pointing at something that no longer accepts the mail. If the partner publishes MTA-STS, review it with the MTA-STS Checker.
| System | Retry period before the 4.4.7 NDR | Delay notice |
|---|---|---|
| Exchange Online | 24 hours of delivery attempts (documented in the 4.4.7 article) | The Exchange Online article doesn't document a delay-notice interval. |
| Exchange Server (on-premises, including hybrid) | Message expiration timeout, default 2 days (Set-TransportService -MessageExpirationTimeOut) | Delay DSN after the delay notification timeout, default 4 hours (Set-TransportService -DelayNotificationTimeOut) |
Microsoft recommends leaving the Exchange Server defaults alone unless Microsoft support or product documentation tells you otherwise. Exchange Server describes 4.4.7 as "Message delayed" while retrying and "Queue expired; Message expired" at the end.
While a message is still retrying, message trace in the Exchange admin center shows its status as Pending (delivery is being attempted or reattempted), and the details can show a Defer event. A final Failed status means it wasn't delivered. Trace results for 10 days or less are available right away; longer ranges, up to 90 days, come as a downloadable report. Resending is fine once the cause is fixed: Microsoft notes that if the receiving server is available, the new message is delivered.
Need SPF, DMARC and MX checks across a whole client list? Run the Bulk SPF / DMARC / MX Auditor on Apify ($2 per 1,000 domains).
Check the destination MX with the MX Record Lookup and DNS Lookup, compare resolvers with the DNS Propagation Checker, review MTA-STS with the MTA-STS Checker, check your SPF with the SPF Record Checker and read the NDR in the Email Header Analyzer. Related guides: Exchange Online 550 5.1.10, Exchange Online 550 5.4.1, Exchange Online 550 5.7.708, unable to get local issuer certificate (incomplete chain) and MTA-STS and TLS-RPT setup.
Exchange Online tried to deliver the message for 24 hours, the destination kept returning temporary errors or couldn't be reached, and the message expired in the queue. Microsoft lists an offline or unreachable destination server, a server that won't accept the message, network timeouts and missing or incorrect MX records as causes.
Microsoft says Exchange Online repeatedly tries to deliver the message over 24 hours, and the sender gets the NDR only after 24 hours of unsuccessful attempts. On-premises Exchange Server uses its own message expiration timeout, 2 days by default.
If it only happens for one destination domain, the problem is most likely on that side: their server, MX record, network or anti-spam settings. Send them the NDR. If many domains or your own on-premises recipients are affected, check SPF, your domain registration, connectors and hybrid routing.
Exchange Online found more than one directory object for the same recipient, for example a mailbox and a mail user or mail contact with the same SMTP or proxy address, or overlapping objects after hybrid sync. Remove the duplicate and let sync finish.
Yes. If the remote server presents only its leaf certificate without intermediates, Exchange Online can't build the chain to a trusted root and reports 450 4.4.317 UntrustedRoot. The remote admin must configure the full certificate chain. Also check partner connectors that require TLS or route to a smart host.
Runbooks, checklists and helpdesk templates: IT Admin Packs on Gumroad.