Adding a Mail Provider? Keep One SPF Record per Domain
Adding a mail sender during a hosting move? Identify the SPF domain, keep one policy per DNS name and verify every sending path without losing existing authorization.
DM
Suppose your website is moving to a new host, but your team’s mailboxes are staying put. You also add a service for order receipts. Its setup screen gives you an SPF record. There’s already one in DNS. Adding the new record underneath looks tidy, but it leaves the receiving mail server with two policies where it expects one.
Keep one SPF TXT record for each domain name that needs an SPF policy. If several services send using that same name, review their authorizations in one policy rather than publishing separate SPF records. First confirm the domain each service actually uses for outgoing mail. Two services don’t always need changes at the same DNS name.
Multiple SPF records for the same domain or subdomain produce permerror, a permanent SPF evaluation error. Microsoft’s SPF setup guidance confirms this rule. The receiver doesn’t combine the records into a larger permission list.
Start with the sender you don’t see in the inbox
The visible From address and the SMTP envelope sender can differ. The latter is also called MAIL FROM or Return Path. Amazon SES explains these identities: its default MAIL FROM uses an Amazon domain, while a custom MAIL FROM uses a configured domain of your own.
Before editing DNS, inspect each service’s current setup instructions and a representative delivered message’s trusted receiving-server authentication results. Record the domain evaluated for SPF. Don’t assume that an address shown as orders@example.com means the service evaluates SPF at example.com; it might use a separate return-path subdomain.
Build a small inventory before touching the record. Google’s SPF setup guide specifically includes website servers, hosted mail, automatic forms and third-party senders in this exercise. A new hosting account is only one possible source.
| Sending path | What to establish | Decision it supports |
|---|---|---|
| Team mailbox | Sending provider and SPF-evaluated domain | Preserve a sender that remains in use |
| Website contact form | Local mail or authenticated external relay | Identify which provider’s instructions apply |
| Order receipts | Default or custom MAIL FROM domain | Update the correct DNS owner name |
This is a planning worksheet, not a report from a real deployment. Fill it with your own evidence. If the form uses the same relay as the mailbox, don’t count it as an unrelated mail platform just because a different application clicked Send.
Two TXT records and two quoted strings are different
An SPF policy lives in a TXT record at its DNS owner name. Other TXT records can coexist there for unrelated purposes. Leave verification tokens alone. RFC 7208’s record-selection rules distinguish SPF records from other TXT data.
Here is an illustrative conflict: two separate records at the same name. The .example include targets are placeholders, not working provider settings.
example.com. IN TXT "v=spf1 include:mail-a.example ~all"
example.com. IN TXT "v=spf1 include:mail-b.example ~all"A single TXT record can contain multiple quoted character strings. The receiver joins those strings without inserting spaces, as RFC 7208 section 3.3 specifies. This next example is one record. The space before the first closing quote matters.
example.com. IN TXT "v=spf1 include:mail-a.example " "include:mail-b.example ~all"The example illustrates record structure, not a complete configuration to paste. Your DNS panel may split long values automatically. Check how that panel represents one record, and use each provider’s exact current values.
Read the answer before trying to repair it
With BIND’s dig available, inspect the exact SPF domain you recorded. Replace example.com. with that name, including a subdomain when appropriate. This read-only command asks your configured resolver for TXT answers.
spf_domain=example.com.
dig "$spf_domain" TXT +noall +comments +answerRead the owner name, response status and record boundaries, rather than counting quote pairs. If caching is in doubt, identify the authoritative nameservers for that owner’s zone through your DNS provider, then query one directly. For a delegated subdomain, use its zone’s servers.
# Replace this with a verified authoritative nameserver.
ns=ns1.dns-provider.example.
dig @"$ns" "$spf_domain" TXT +norecurse +noall +comments +answerCheck for a successful response and the aa flag indicating an authoritative answer. Repeat for the other authoritative servers. A timeout, refusal or non-authoritative response doesn’t establish the intended record set. The BIND manual documents server selection and these dig options.
Review one replacement policy, not a blind concatenation
Keep a copy of the current value and its TTL. Ask the DNS or mail owner to review one replacement policy for that exact domain, accounting for every sender still needed during the transition. Keep the existing failure qualifier unless a deliberate policy change has been agreed. More permissive authorization is not a useful substitute for finding the missing sender.
Merging records can expose another problem: SPF evaluation has a ten-lookup limit for mechanisms and modifiers that trigger DNS lookups. Nested includes count too, so counting only the visible include: terms is insufficient. Microsoft explains the lookup limit and nested evaluation. Validate the complete proposed policy before applying it.
Use your DNS provider’s supported edit or record-set replacement workflow so the intended final state has one SPF policy, without leaving a duplicate behind. If you can’t tell how the panel applies the change, have the DNS owner handle it. Don’t replace unrelated MX, DKIM or verification records just to solve this one conflict.
Recheck the authoritative answers after the change, then check ordinary resolver answers. Existing caches can retain an earlier value until its cached TTL expires; Cloudflare’s TTL reference explains the cache behavior. A correct authoritative answer and an old recursive answer call for a different response from two conflicting authoritative records.
A clean DNS answer is the start of the mail check
Arrange a controlled test through each sending path in your inventory: mailbox, form and receipt service. At the receiving system, inspect the trusted authentication result and the domain it evaluated. Record which path passed, failed or wasn’t tested. One successful mailbox message doesn’t establish that the website’s receipts use the same sender.
SPF checks sending authorization; it doesn’t guarantee inbox placement or successful delivery. DKIM, DMARC alignment, receiver policy and other delivery factors remain separate checks. If the error changes after the duplicate is removed, investigate the new result rather than treating one SPF record as proof that the whole mail setup is correct. Microsoft’s authentication guidance treats SPF, DKIM and DMARC together.
For a hosting move, carry this sender inventory alongside the WordPress migration checklist. If the website itself reaches different destinations, the A and AAAA cutover guide addresses that separate routing problem.
Nest is Krova Cloud’s managed hosting product for websites and email. When evaluating a move, confirm the mailboxes, sending paths and provider-supplied DNS values for your account. This article doesn’t prescribe a Krova SPF include or claim that changing website hosting automatically moves all outgoing mail.
Plan the website and email move together
Explore Nest’s managed hosting workflow and map which mail services will stay, move or be added before changing DNS.
Creation context: this article and cover were produced with AI assistance using the linked primary documentation, checked on 17 November 2026 IST. The scenario, worksheet and DNS examples are illustrative. No DNS changes, test emails, customer incident or personal testing by the bylined author are claimed.




