[ Headers & delivery errors ]

Mailbox-full responses and their effect on retry handling

Keep a capacity refusal separate from sender authentication, and let its actual SMTP class and queue state govern the next attempt.

By the EmailAuth research team · 3 min read

[ Key points ]

  • Keep a capacity refusal separate from sender authentication, and let its actual SMTP class and queue state govern the next attempt.
  • Decide whether a replacement is needed
On this page

A recipient whose mailbox is full may still be a valid and willing recipient. The immediate task is to determine whether the original message remains queued, whether the receiving system considers the failure temporary, and whether the message will still be useful when space becomes available. Adding an SPF include will not create mailbox capacity.

Read the class and the capacity diagnostic

RFC 3463 defines X.2.2 as mailbox full and says it should be used as a persistent transient failure. The first component supplies the success, persistent transient failure or permanent failure class. Preserve the complete SMTP reply as well as the enhanced status: a dashboard's “soft bounce” label may hide the actual response and transaction stage. RFC 3463 section 3.3.

A 4xx reply normally leaves responsibility for retry with the sending system. A 5xx reply indicates permanent failure for that attempt and should not trigger routine repeated attempts without intervention. If the three-digit SMTP class and enhanced class disagree, retain both and investigate the server or parser behavior rather than silently rewriting the evidence. The underlying storage condition can later change even when a particular attempt has permanently failed.

Establish which message remains queued

Capture the sender's queue identifier, original submission time, failed envelope recipient, last response and next scheduled attempt. A multi-recipient message can have one deferred recipient while others have already been accepted. Resending to the complete visible recipient list can therefore duplicate messages for people unaffected by the capacity problem.

Check whether the sender still owns the delivery attempt or whether a downstream relay accepted responsibility and later generated a notification. A delay warning is not necessarily a final failure. Use the queue record and recipient-specific notification fields together. Keep original submission time separate from notification arrival time, especially when the sender reports that the failure happened “just now.”

Work through a hypothetical delayed message

Suppose queue item Q378 receives 452 4.2.2 for one recipient. The recipient's administrator later confirms an approved capacity repair. Before creating a replacement, check Q378: if it still awaits a scheduled retry, follow that queue; if its later attempt has already received final DATA acceptance, an old warning is not a reason to send another copy.

If Q378 expired before capacity was restored, retain its failed outcome and identify any authorized replacement separately. The repair does not retroactively deliver the expired item. Other recipients may already have succeeded, so avoid resending the entire visible recipient list. These cases are hypothetical, not a live mailbox test, retained queue or measured recovery.

Resolve the capacity issue without multiplying mail

For an organization-controlled recipient, the mailbox owner or administrator should review quota, retention and available storage through approved procedures. Do not delete user data merely to clear a delivery alert. For an external recipient, use an established alternative channel when the business message is urgent, and let that person resolve their own mailbox conditions.

After remediation, first check whether the original queued item has already succeeded. Initiate a new authorized submission only when its relationship to the original attempt is understood. Respect the sender's documented retry and expiry policy rather than hammering the receiver. RFC 5321 describes retry intervals and configurable give-up behavior; its general guidance is not a promise that every provider retains mail for the same period. RFC 5321 section 4.5.4.1.

Close the incident with the recipient disposition, capacity correction and final delivery evidence. SMTP acceptance establishes a handoff, not proof that the recipient read the message. If the original has expired, record that fact separately from any successfully submitted replacement.

Decide whether a replacement is needed

  • Original still queued: check next retry and usefulness before making another copy.
  • Original accepted after repair: do not resend merely because an old warning arrived.
  • Original expired: record that outcome separately; authorize a replacement only if still useful.
  • Mixed recipients: limit any new send to the intended unresolved recipients.
  • Capacity repair: retain owner-approved evidence without exposing mailbox contents.

[ Takeaway ]

Keep a capacity refusal separate from sender authentication, and let its actual SMTP class and queue state govern the next attempt.

[ By role ]

What this means for you

Choose your role for practical next steps.

IT / Mail Admin

Follow the recipient queue state

The failed attempt may still be active.

Next steps

  1. Capture queue age and response.
  2. Confirm receiving capacity remediation.
  3. Check the original before resubmitting.

CISO / SecOps

Preserve retention controls

Delivery pressure does not authorize indiscriminate deletion.

Next steps

  1. Review storage-change ownership.
  2. Protect diagnostic recipient data.
  3. Keep sender authentication intact.

Founder / Non-technical

Keep urgent communication moving

Capacity problems can delay a critical workflow without invalidating the contact.

Next steps

  1. Identify time-sensitive messages.
  2. Use an approved alternative channel.
  3. Avoid duplicate business actions.

[ FAQs ]

Frequently asked questions

Does mailbox full mean the address is invalid?

Not necessarily. Capacity and destination-address validity are separate conditions.

Which enhanced status identifies mailbox full?

RFC 3463 defines X.2.2; retain its class and the complete SMTP diagnostic.

Should every capacity response be retried?

Follow the actual response class and queue policy. A permanent refusal needs intervention rather than unchanged routine retries.

Could only one recipient be deferred?

Yes. Track each envelope recipient so a resend does not duplicate successful deliveries.

Does a delay warning mean the queue is empty?

No. Confirm the current queue state before creating another submission.

Was the recovery measured on a real mailbox?

No. It is a hypothetical queue-state example. A real incident needs the current queue record, approved capacity repair and per-recipient outcome.

Who should remove mailbox content?

The authorized owner or administrator under applicable retention and deletion procedures.

What confirms recovery?

The original or replacement attempt has a documented final outcome, with its identity and recipient scope preserved.

[ Sources ]

Sources

  1. RFC 5321 SMTP

    Sections §3.3 · §4.1.1.1 · §4.2 · §4.5.3.2 · §4.5.4.1 · §5 · §6.1

  2. RFC 3463 Enhanced Mail System Status Codes

    Sections §3.2 · §3.3 · §3.4 · §3.5

  3. RFC 3464 Extensible Message Format for Delivery Status Notifications

    Sections §2.2 · §2.3

[ Glossary ]

Glossary

Mailbox quota
A configured limit on mailbox storage or usage.
Queue expiry
The point at which a sending system stops trying an unresolved delivery.
Deferred recipient
A recipient whose delivery remains pending after a temporary condition.

Enterprise provenance

Why top financial institutions and governments retired their email security dashboards.

Traditional email authentication platforms sell you a dashboard and ask your engineers to do the labour. In high-stakes environments, manual oversight means delayed enforcement, configuration drift and unmonitored shadow domains.

EmailAuth is the agentic email provenance platform. Our agents enforce p=reject, keep third-party senders aligned and move your domains to full RFC 9989 compliance, the new DMARC standard,, with human-in-the-loop auditability.

Governments Large enterprises Tier-1 financial institutions

SaaS · Dedicated tenancy · Air-gapped on-prem · BYOK encryption

Knowledge Base