Information checked: 21 July 2026.
Email problems are easier to resolve when you separate message submission, DNS authentication, delivery, filtering and the recipient’s mailbox. Repeatedly changing DNS or resending the same message can make the incident harder to understand.
In brief: Follow one message from the sending application to the recipient mailbox and make changes only where the evidence fails.
Start with one precise example
- Exact sender and recipient addresses.
- Time sent, including time zone.
- Message ID or full headers where available.
- Bounce code and complete error text.
- Whether the problem affects one recipient, one domain or everyone.
- Any DNS, password, provider or website change made shortly beforehand.
Troubleshoot by stage
| Stage | Evidence | Typical problem |
|---|---|---|
| Application submission | Website or application mail log | Wrong credentials, blocked port or incorrect From address |
| Sending provider | Provider message trace | Rate limit, policy rejection or account suspension |
| DNS authentication | SPF, DKIM and DMARC results in headers | Missing, duplicated or misaligned records |
| Recipient gateway | Bounce and remote response | Reputation, attachment or policy block |
| Mailbox | Rules, quarantine and storage status | Message delivered but hidden or rejected locally |
Protect the website form workflow
A contact form should record a submission safely before sending notification. Use an address on your own domain in the From field and place the visitor’s address in Reply-To where appropriate. Do not rely on a single unlogged PHP mail call as the only record of an enquiry.
Avoid destructive fixes
- Do not replace MX, SPF or DKIM records from an unrelated tutorial.
- Do not publish more than one SPF record for the same hostname.
- Do not disable spam or authentication controls globally to make one test pass.
- Do not delete the old provider until all historical mail and routing are verified.
Close the incident
Document the cause, exact change, affected period and messages that may need to be resent. Add monitoring or a configuration record so the next person can distinguish a provider outage from a local mailbox problem.
Example: only website messages fail
Staff can send from Outlook, but the contact form fails after a password rotation. This usually points to the application submission stage rather than MX or recipient filtering. Test the same credentials through the configured SMTP service, inspect the application log and avoid changing domain-wide DNS records until evidence shows an authentication problem.
What the business should retain
- One complete failed-message example
- Full headers or bounce text
- Current DNS authentication records
- Website mail configuration owner
- Incident cause and corrective action
Sources and date checked
Technical and product information was checked against the following primary sources on 21 July 2026. Plans, interfaces and provider terms can change, so recheck them before purchase or migration.
Keep the decision under your control
Retain the relevant accounts, source material, supplier terms and recovery information. Recheck changing prices, interfaces and rules before acting.