Error Details
| Fact | Value | Source | Note |
|---|---|---|---|
| Exact error string | 421 4.3.2 SERVICE.UNAVAILABLE | Exchange Online message trace | -> Deferred status |
| Affected build range | Microsoft 365 / Exchange Online | Microsoft documentation | |
| Registry key or OMA-URI involved | |||
| Relevant service/endpoint URL | admin.microsoft.com > Health > Service health | Microsoft 365 Admin Center | -> Check for active incidents |
| Log line signature | 421 4.3.2; Deferred status | Exchange Online message trace logs |
Log Files to Check
C:\Program Files\Microsoft Monitoring Agent\Agent\HealthServiceLogs\HealthService.log/var/log/exchange/healthservice.log
What This Error Actually Means
The error code "421 4.3.2 SERVICE.UNAVAILABLE" is a transient SMTP response that indicates the receiving mail server, specifically an Exchange Online Protection (EOP) server or an on-premises hybrid connector configured to communicate with EOP, is temporarily unable to accept messages from the sending mail transfer agent (MTA). This error is raised by the EOP service when it encounters conditions such as excessive traffic from a particular IP address or domain, maintenance windows, or issues related to sender reputation. The "421" response code signifies that the server will not accept any more connections at this time, while "SERVICE.UNAVAILABLE" indicates the specific reason for rejection.Affected Versions and Scope
This error is relevant to all Microsoft 365 tenants using Exchange Online as their email service. It affects both on-premises hybrid environments that are synchronized with Exchange Online and cloud-only deployments. The issue does not impact standalone on-premises Exchange Server installations unless they are configured to communicate directly with EOP for spam filtering or other services.Where To Look First: Logs and Event IDs
To diagnose the root cause of this error, start by examining logs from both Exchange Online and your on-premises environment. In Exchange Online, you can find relevant entries in the message trace logs which show a 'Deferred' status with the exact log line signature "421 4.3.2". Additionally, check the HealthService.log file located atC:\Program Files\Microsoft Monitoring Agent\Agent\HealthServiceLogs\ on Windows or /var/log/exchange/healthservice.log on Linux for any related events.
For on-premises environments, inspect your mail server logs and look for repeated 421 responses from EOP servers identified by the domain "*.mail.protection.outlook.com". Event IDs to pay attention to include 3005, which typically indicates a problem with SMTP communication, and 9518, which can indicate issues related to connector configuration or excessive traffic.
Root Causes
EOP Throttling Due to Excessive Traffic
When the sending IP address or domain sends email at a rate above what is permitted by EOP, it may trigger throttling mechanisms that temporarily reject incoming messages. This can be identified in logs showing repeated 421 responses from EOP servers and in the Exchange Online message trace with 'Deferred' status.EOP Maintenance Window
EOP occasionally undergoes maintenance periods during which services might not be fully operational, leading to temporary unavailability. While rare and typically short-lived (minutes), this can still cause transient 421 errors. Check the Microsoft 365 Service Health Dashboard for any active incidents affecting Exchange Online.High Composite Spam Confidence Level (CSCL) Reputation
If a sending IP has a high CSCL, it may be temporarily deferred by EOP to prevent spam from reaching recipients' inboxes. This can result in the 421 error code being returned until the sender's reputation improves.On-Premises Hybrid Connector Misconfiguration
Misconfigured hybrid connectors or excessive retry storms caused by incorrect settings on the connector can lead to repeated 421 errors as EOP rejects messages due to perceived overloading. Review your connector configurations and ensure they are properly set up for optimal performance.Fix Procedure
- Wait for Automatic Resolution: Most transient issues resolve themselves within 15-60 minutes, so the first step is often to wait.
- Check Service Health Dashboard: Visit
admin.microsoft.com > Health > Service healthand review any active incidents affecting Exchange Online. - Review Sending Patterns: Use PowerShell cmdlets like
Get-MessageTrackingLog -ResultSize Unlimited | Where-Object {$_.EventID -eq "421"}to analyze sending patterns and identify potential issues with excessive traffic. - Implement IP Throttling on Premises: If the issue is due to high volume from an on-premises relay, configure IP throttling settings using PowerShell cmdlets such as
Set-TransportConfig -SmtpClientAuthenticationDisabled $false. - Check Sender Support Portal: For flagged sending IPs, visit the Microsoft Sender Support portal at sendersupport.olc.protection.outlook.com for guidance.
- Implement SPF, DKIM, and DMARC: Improve sender reputation by configuring these protocols using the Exchange admin center (EAC) > Mail flow > Rules/Connectors/Accepted domains.
Error Code Variants
| Error Code | Description | Distinguishing Features |
|---|---|---|
| 451 4.3.0 | Local Error Processing Halted | This error indicates a local processing issue rather than an external service being unavailable. |
| 421 4.7.0 | Service Not Available - General Failure | Indicates a more generic failure of the mail server, not specifically related to EOP throttling or maintenance. |
Verification
To verify that the issue has been resolved, monitor Exchange Online message traces for messages transitioning from 'Deferred' status back to 'Delivered'. Additionally, check the HealthService.log file for entries indicating successful reconnection attempts. In PowerShell, you can useGet-MessageTrace -StartDate (Get-Date).AddHours(-2) -EndDate (Get-Date) to track recent message delivery statuses.
Prevention
Implementing robust sender reputation management practices is crucial for preventing recurrence of this error. This includes configuring SPF, DKIM, and DMARC records correctly, monitoring sending patterns closely, and ensuring that hybrid connectors are properly configured to avoid excessive retries or traffic spikes. Regularly review the Microsoft 365 Service Health Dashboard for any maintenance windows or service issues that could affect your environment.Be cautious when making changes to connector configurations as incorrect settings can disrupt email flow and impact enrolled devices.
Use PowerShell cmdlets like
Get-RemoteDomain | Select-Object DomainName,SmtpRejectMessagesFromGroups for quick diagnostics of remote domain settings affecting message delivery.