Skip to content

Est. 2011ยทMicrosoft Partner ID 7033487ยทDelivery under 3 minยทSupport 7 days a week

Your vault is empty.

Free Fix 550 5.1.1

550 5.1.1 and 550 5.1.10: Exchange cannot find the recipient mailbox

10 min read Updated October 5, 2026 Exchange Server

Fix it now

Exchange accepted responsibility for the domain, looked the address up and found nothing deliverable. Before touching the directory, read the failed address in the diagnostic section: if it begins IMCEAEX- rather than looking like an email address, this is a stale Outlook AutoComplete entry and the repair is different.

Run in the Exchange Management Shell, or after Connect-ExchangeOnline for a cloud mailbox

Get-Recipient -Identity user@contoso.com | fl Name,RecipientType,PrimarySmtpAddress,EmailAddresses
Get-Recipient -ResultSize Unlimited -Filter "DisplayName -like '*Surname*'" | fl Name,PrimarySmtpAddress
  1. If the failed address starts with IMCEAEX-, the sender is replying from a cached entry that points at an old legacyExchangeDN. Rebuild that string as an X500 address and add it to the recipient; the sender’s own address book is not the fix.
  2. If it is an ordinary address and nothing comes back, search by name instead. Most of these bounces are a genuine typo or an alias that was retired.
  3. If an object comes back but RecipientType is MailUser or MailContact, there is no mailbox behind the address. Confirm with Get-Mailbox on the same identity.
  4. If the mailbox was moved to Microsoft 365, check that the on-premises object carries the routing address alias@domain.mail.onmicrosoft.com. A move that left it unstamped bounces exactly like this.

Nothing here costs money. Every repair is a directory change, and any advice that opens by selling you something for a 5.1.1 is wrong.

If a test message now lands, you are done. If not, the next section covers what happens between accepted domain and mailbox, and which of the five ways a recipient goes missing you have.

Why it happens

When a message arrives for a domain the organisation is authoritative for, Exchange takes responsibility for it and then resolves the address to an object. That resolution matches on proxy addresses, not on names. Every SMTP address stamped on a recipient, primary or secondary, is a valid way to reach it, and an address that is not on the object resolves to nothing however obviously it belongs to the person.

The reason to start with the IMCEAEX- test is that Outlook does not send to the address you see. Inside the organisation it sends to the recipient’s legacyExchangeDN, cached in AutoComplete. Change that attribute – by migrating, recreating an account, or moving a mailbox between organisations – and every cached entry pointing at the old value fails, while the address itself resolves perfectly for anyone who types it fresh. That is the case Microsoft documents first, and a directory lookup cannot see it.

The rest are ordinary directory faults, and none of them is a spam decision, a size limit or a routing problem. The repair belongs to the recipient object rather than to transport.

A stale legacyExchangeDN reached through AutoComplete

You have this one if The failed recipient in the NDR is a long IMCEAEX- string ending in something that looks like a domain, not an email address. Typing the address fresh works.

  1. Rebuild the string as an X500 address using the substitutions in the reference section below.
  2. Add it to the recipient: Set-Mailbox user -EmailAddresses @{add='X500:/O=.../CN=...'}
  3. Have the sender retry. The cached entry now resolves again.

Clearing the sender’s AutoComplete entry also works for that one sender. Adding the X500 address fixes it for everyone who ever cached the old value, which is why it is the documented remedy.

A mailbox moved to Microsoft 365 without its routing address

You have this one if The mailbox moved recently, the on-premises object still exists, and mail from inside the organisation bounces while external mail arrives.

  1. Open the on-premises object and check its email addresses for alias@domain.mail.onmicrosoft.com.
  2. Add it if missing, and set it as the routing address.
  3. In the Exchange Admin Center this is Recipients, then Mailboxes, then the user, then Email Address, then Add.

The address was never on the object

You have this one if A lookup on the exact address returns nothing, but a search by surname finds the person under a different address.

  1. Send to the address the directory search returned.
  2. If the old address must keep working, add it as a secondary proxy: Set-Mailbox user -EmailAddresses @{Add='old.alias@contoso.com'}.
  3. Check whether an email address policy owns the object and will strip it again. If so, add the address to the policy instead.

Adding a proxy address does not create a second mailbox and costs nothing. The primary address, which decides what recipients see in the From line, is unchanged.

The object is mail-enabled but has no mailbox

You have this one if Get-Recipient returns a MailUser or MailContact; Get-Mailbox fails on the same identity.

  1. For a contact or mail user, read where it forwards: Get-MailContact <name> | fl ExternalEmailAddress. If that target is dead, the bounce is from the far end.
  2. For someone who should have a mailbox, create or reconnect one rather than adding another proxy address.
  3. Check for a disconnected mailbox in the database first, so history survives.

One side of a hybrid organisation has no object for the user

You have this one if 550 5.1.10 rather than 5.1.1, for someone who demonstrably exists.

  1. Establish where the user was created. Microsoft’s documented case is a user created in Microsoft 365, leaving the on-premises organisation with nothing to reference.
  2. Create the matching on-premises object with New-RemoteMailbox, using the same name, alias and UPN as the cloud mailbox.
  3. Copy the ExchangeGuid across: read it with Get-Mailbox <name> | fl ExchangeGuid in Exchange Online, then Set-RemoteMailbox <name> -ExchangeGuid <guid> on premises.
  4. Run Start-ADSyncSyncCycle -PolicyType Delta and confirm the cloud object shows as synced with Active Directory.

Change the attribute on whichever side is authoritative. A synchronised object cannot be repaired by editing the cloud copy, and a cloud-only object cannot be repaired on premises.

Full reference

Rebuilding an IMCEAEX string into an X500 address

The encoding is mechanical, so it converts by hand in under a minute. Work through the substitutions, then strip the wrapper.

In the NDR In the X500 address
_ /
+20 a space
+28 (
+29 )
+2C ,
+2E .
+5F _
The IMCEAEX- prefix removed
The trailing @domain removed

Prefix the result with X500: and add it to the recipient with Set-Mailbox. The address becomes another way to reach the same object, exactly as an old SMTP alias would, and no mail is rerouted by adding it.

Which shell you are in matters

Task On-premises Exchange Exchange Online
Look up a recipient Get-Recipient Get-Recipient after Connect-ExchangeOnline
Trace a message Get-MessageTrackingLog Get-MessageTraceV2
Find disconnected mailboxes Get-MailboxStatistics -Database <db> Not applicable
Force a directory sync Start-ADSyncSyncCycle -PolicyType Delta on the sync server Same, run on premises
Create the on-premises half of a cloud mailbox New-RemoteMailbox Not applicable

Get-MessageTrackingLog is available only in on-premises Exchange. Reaching for it in an Exchange Online session is one of the more common ways a hybrid investigation stalls; the cloud equivalent is Get-MessageTraceV2.

When the recipient object checks out

  • Confirm only one object holds the address. Two objects claiming it produce 5.1.4 and intermittent delivery, and stale contacts left behind by an old migration are the usual duplicate.
  • Allow for replication. On premises, a change made on one domain controller is not visible everywhere immediately, and a retest inside that window proves nothing.
  • Check whether an email address policy owns the object. A policy will put addresses back the way it wants them, quietly, after you have finished editing.
  • Check the accepted domain type. If the domain is authoritative, unknown recipients are refused; if it is internal relay, they are passed onward and the bounce may be coming from the system behind you.
  • Read the generating server line in the NDR. It tells you which organisation actually made the decision, which is not always the one you assumed.

On turning recipient checking off

It is technically possible to make an accepted domain non-authoritative so that unknown recipients are relayed onward instead of refused, and it will stop the bounces. It is a mail flow decision with consequences rather than a fix: clean rejections become backscatter from whatever sits behind you, and an address space that accepts everything is convenient for anyone probing you for valid recipients. That is engineering judgement rather than a documented Microsoft position, but it is the reason the recipient object is the better place to spend the time.

Every code this article covers

Code What it points at Source
550 5.1.1 Bad destination mailbox address: the address did not resolve to a deliverable object. Microsoft’s documented causes are a changed legacyExchangeDN reached through a cached entry, and a moved mailbox with no routing address Microsoft Learn
550 5.1.10 Recipient not found. Microsoft’s documented case is a user created in Microsoft 365 with no matching on-premises object to reference Microsoft Learn
550 5.4.1 Relay access denied, or recipient address rejected: access denied. What an unknown recipient looks like when the receiving system will not relay for it Microsoft Learn
550 5.1.6 Destination mailbox has moved, no forwarding address: the address was valid once and mail is no longer accepted for it RFC 3463
550 5.1.3 Bad destination mailbox address syntax: the address was refused as malformed before any directory lookup RFC 3463
550 5.1.4 Destination mailbox address ambiguous: the address matches more than one recipient on the destination system RFC 3463
550 5.1.2 Bad destination system address: the domain part does not exist or cannot accept mail RFC 3463

Confirm the fix worked

  1. Get-Recipient -Identity user@contoso.com returns exactly one object, and Get-Mailbox on the same identity succeeds.
  2. A test message reaches the mailbox: check Get-MessageTrackingLog on premises, or Get-MessageTraceV2 in Exchange Online.
  3. The original sender retries from Outlook without clearing anything, which is what proves the cached entry resolves again.
  4. In a hybrid organisation, the same object comes back from both the on-premises shell and the Exchange Online session.
  5. The recipient confirms the message arrived in the mailbox you expected, not in a shared or forwarded copy.

Questions people ask about this

Does fixing this cost anything?

No. Every repair here is a directory change: adding a proxy or X500 address, stamping a routing address, reconnecting a mailbox or running a sync cycle. No licence, product or subscription is involved.

Why does the address work when I type it but not when I reply?

Because Outlook does not send to the address you see. Inside the organisation it uses the recipient’s legacyExchangeDN, cached in AutoComplete, and if that value has changed the cached entry points at something that no longer exists. Adding the old value back as an X500 address repairs it for every sender at once.

Why did the message bounce hours after I sent it?

A receiving system that accepts on the strength of the domain and only then discovers the recipient does not exist has to generate the bounce after acceptance. A system that checks recipients while the sending server is still connected refuses immediately instead, which is why some of these arrive in seconds and others overnight.

The person definitely works there. Why is their address unknown?

Because addresses are attributes, not identities. A rename, a merger, a change of email address policy or a migration can each leave a working mailbox reachable only at an address you do not have. Ask them to send you a message and reply to whatever appears.

I get 5.1.10 for a mailbox I can see in the admin centre. What is wrong?

Look at the other side. Microsoft documents this as a user created in Microsoft 365 with no on-premises object behind it, so queries from the on-premises organisation find nothing. The fix is to create the matching on-premises object with a matching ExchangeGuid and let it sync.

Related error codes

Was this article helpful?

Your feedback helps us improve our documentation.

Related articles

Free Fix 554 5.6.0 invalid content: malformed messages Exchange refuses to accept Free Fix 0x80090327 and 0x800B0109: Exchange TLS fails on certificate validation License Error Event ID 9518 with error -1808: the log volume filled and the store dismounted Free Fix HCW8064: the Hybrid Configuration Wizard finishes but OAuth is not set up
โ† Back to Knowledge Base