Agent view · Guides · 2026-09-30
How to move DMARC from p=none to p=quarantine without losing email
This is the response this server returns to a client that asks for Accept: text/markdown at https://crawlcheck.io/guides/how-to-move-dmarc-from-none-to-quarantine-without-losing-mail: fetched when you opened this page, shown byte for byte. An assistant, a crawler that negotiates markdown, or anything reading our feed gets exactly this. The markdown leaves signed (RFC 9421, key in our key directory), so a crawler can prove the page it read is the page this server sent; the check is shown below.
| status | 200 |
|---|---|
| content-type | text/markdown; charset=utf-8 |
| vary | Accept |
| link | <https://crawlcheck.io/llms.txt>; rel="describedby"; type="text/plain", <https://crawlcheck.io/.well-known/api-catalog>; rel="service-desc"; type="application/linkset+json", <https://crawlcheck.io/guides/how-to-move-dmarc-from-none-to-quarantine-without-losing-mail>; rel="cite-as", <https://crawlcheck.io/.well-known/http-message-signatures-directory>; rel="http-message-signatures-directory" |
| cache-control | public, max-age=900, stale-while-revalidate=2400 |
| bytes | 21,438 |
| words | 3,192 |
| sha-256 | afc471764727cee2d498ee74dd990400a25680f599c415dcb735e24303d2200f |
| repr-digest | sha-256=:r8RxdkcnzuLUmO503ZkEAKJWgPWZxBXctzXiQwPSIA8=: ← matches the body shown |
| signature-input | sig1=("@authority";req "@path";req "repr-digest" "content-type");created=1795182066;expires=1795185666;keyid="1hHmE3fo1rg5Ig70wuFtJ5wBc7G-sUTgUJkxyPWFRps";alg="ed25519";nonce="AVPgDFrMDxhjQidt45jonQ";tag="crawlcheck-page" |
| signature | sig1=:d79npN8hwpDHRXOcc4YYIhwEnjFejaHI33a5Co5Hv/w8ng581EOQi+08xfg0G+R1HJuXFsNihiWY07nhL61OCw==: |
| verified | yes — signature by key 1hHmE3fo1rg5Ig70… from https://crawlcheck.io/.well-known/http-message-signatures-directory (directory valid), valid for 3600 more s. Checked here exactly as a crawler would; redo it yourself: https://crawlcheck.io/api/tool/wba |
Guides · 2026-09-30 · By [VSNARY | Emmanuel Orta](/author) · 4 views # How to move DMARC from p=none to p=quarantine without losing email The sequence for tightening DMARC safely: inventory every sender, prove alignment from received headers, create the report mailbox first, then raise the policy in stages. DMARC passes when SPF or DKIM passes on a domain aligned with the visible From address. To move from p=none to p=quarantine without losing mail: list every system that sends as your domain, read the Authentication-Results header on a real message from each, get DKIM aligned on all of them (SPF alone fails on forwarding), create the rua report mailbox before the record names it, then publish quarantine at pct=25 and raise it as the aggregate reports stay clean. Set sp= deliberately. Check your own site free Check my site About a minute. First scan free with your email, nothing installed. On this page - What does DMARC actually evaluate? - Step one: inventory every sender - Step two: prove each one passes, from a received message - Why is SPF alone not enough? - Step three: give the reports somewhere to land - How do you read an aggregate report? - Step four: how do you raise the policy in stages? - What will quarantine not fix? - Checklist - The SPF lookup limit A [DMARC](/glossary/dmarc) record with `p=none` is a monitoring instruction: it asks receiving mail servers to report what they see and to deliver everything anyway. Most domains that have a record stop there, because the next step, `p=quarantine`, is the one that can send your own legitimate mail to spam if anything is misconfigured. This guide is the sequence for making that move safely, written from doing it on crawlcheck.io in September 2026, where two separate sending systems both had to pass before the policy could tighten, and updated for the DMARC revision published as RFC 9989 in May 2026. **Key points** - DMARC passes when [SPF](/glossary/spf) or [DKIM](/glossary/dkim) passes on a domain aligned with the visible From address. - Inventory every sender, then prove each one from the `Authentication-Results` header of a real delivered message. - Get DKIM aligned on every sender: SPF fails whenever mail is forwarded, and aligned DKIM survives it. - Create the `rua=` report mailbox before the record names it. - RFC 9989 (May 2026) removes `pct` and adds a `t=y` testing flag that applies the next less strict policy; it is not an off switch. ## What does DMARC actually evaluate? # DMARC adds no new authentication. It asks whether SPF or DKIM passed for a domain that matches the visible From address. DMARC does not authenticate anything itself. It reads the results SPF and DKIM already produced and asks about alignment. Did SPF pass, and is the domain SPF checked (the envelope sender) the same organisational domain as the visible From address? Did DKIM pass, and is the `d=` domain in the signature the same organisational domain as the From address? If either pair passes and aligns, DMARC passes. If neither does, the policy applies: `none` delivers, `quarantine` asks the receiver to treat the message as suspicious, `reject` asks it to refuse. The most common cause of legitimate mail failing under `quarantine` is a sender that passes on a domain that does not align: a marketing platform that signs with its own domain, or a form tool that uses its own envelope sender. ## Step one: inventory every sender # Before touching the policy, list every system that sends mail with your domain in the From address. Mailboxes are the obvious one. The rest hide: a transactional sender inside your web application, a form tool, a CRM, a support desk, a newsletter platform, a monitoring service that emails alerts, a billing provider. On crawlcheck.io there were two: the Google Workspace mailboxes, and the site’s own application mail. They use different SPF includes and different DKIM selectors, and both had to align independently. ## Step two: prove each one passes, from a received message # The reliable proof is not the DNS record; it is the `Authentication-Results` header on a message each sender actually delivered. Send one message from each system to a mailbox you control, open the raw source, and read the line (shortened here, with the application’s selector and bounce subdomain shown as placeholders): ``` Authentication-Results: mx.google.com; dkim=pass header.i=@crawlcheck.io header.s=<app-selector> header.b=…; spf=pass (google.com: domain of …@<bounce-subdomain>.crawlcheck.io designates … as permitted sender); dmarc=pass (p=QUARANTINE sp=QUARANTINE dis=NONE) header.from=crawlcheck.io ``` For the application mail, DKIM passed on `crawlcheck.io` and SPF passed on a bounce subdomain of `crawlcheck.io`, which aligns at the organisational level. For the Workspace mailboxes the same line first showed SPF passing and no DKIM at all, because the Workspace DKIM key had never been generated. That is a pass today and a single point of failure. | Sender | SPF | DKIM | Aligned | Under quarantine | | --- | --- | --- | --- | --- | | Workspace mailbox, before DKIM | pass (include:_spf.google.com) | none | SPF only | passes, fails if forwarded | | Workspace mailbox, after DKIM | pass | pass, s=google, d=crawlcheck.io | both | passes | | Application mail | pass on a bounce subdomain | pass, own selector, d=crawlcheck.io | both | passes | | a form tool signing its own domain | pass on its domain | pass, d=formtool.example | neither | quarantined | ## Why is SPF alone not enough? # Mailing lists, alias services and forwarding rules all re-send mail from their own servers. Aligned DKIM is what keeps that mail passing once the policy tightens. A forwarding server re-sends the message from its own address, which is not in your SPF record, so SPF fails at the final receiver. The DKIM signature is computed over the headers and body, which forwarding normally leaves alone, so it still verifies. That is why the safe precondition for quarantine is aligned DKIM on every sender, not SPF on every sender. ## Step three: give the reports somewhere to land # Aggregate reports are the only view of senders you forgot to list. They are worth nothing if the address in rua= does not receive mail. The record names a `rua=` address for aggregate reports. Receivers send one XML report per day per receiver naming every source IP that sent as your domain and how each fared; this is the only view you have of senders you forgot in step one. On crawlcheck.io the record pointed reports at `dmarc@crawlcheck.io` before that alias existed, so the reports bounced for the whole period they were most needed. Create the mailbox or alias first, send a test to it, and only then publish the record that names it. ## How do you read an aggregate report? # Each report is a small XML file. The part that matters is one `record` per sending source: ``` <record> <row> <source_ip>203.0.113.7</source_ip><count>42</count> <policy_evaluated><disposition>none</disposition> <dkim>fail</dkim><spf>pass</spf></policy_evaluated> </row> <identifiers><header_from>example.com</header_from></identifiers> </record> ``` Read it as: 42 messages claiming to be from your domain came from 203.0.113.7; DKIM did not align, SPF did, so DMARC passed on SPF alone. Look up who owns the address. If it is a system you use, it is a sender to fix before tightening, because the first forwarded message from it will fail. If you do not recognise it at all, it is either a sender nobody told you about or someone forging your domain, and the report is how you find out which. Reports are compressed and arrive per receiver, so a parser or a reporting service saves time once volume grows; reading a few by hand first teaches you what the numbers mean. ## Step four: how do you raise the policy in stages? # Under RFC 9989 (May 2026) the t=y testing flag replaces pct: it applies the next less strict policy, so p=reject; t=y is enforced as quarantine. It is not an off switch. The original specification staged enforcement with `pct=`, applying the policy to a percentage of failing messages. RFC 9989, which replaced RFC 7489 in May 2026, removes `pct` because receivers applied it inconsistently, and adds a testing flag. With `t=y`, receivers apply the next less strict policy: `p=reject; t=y` is enforced as quarantine, and `p=quarantine; t=y` as none. Receivers may still honour `pct` during the transition, but new records should use the new tag. Note what `t=y` is not: it is not the old `pct=0` off switch, and treating it as one loosens a domain by a full level. | Record | What failing mail gets | Use it when | | --- | --- | --- | | `p=none` | delivered; reported | you are still listing senders | | `p=quarantine; t=y` | delivered (applied as none) | you want intermediaries to see the intended policy before enforcing it | | `p=quarantine` | spam folder | every known sender aligns DKIM and the reports are clean | | `p=reject; t=y` | spam folder (applied as quarantine) | quarantine has run clean and you are preparing for reject | | `p=reject` | refused | reports show no legitimate source failing over a full business cycle | Two more tags deserve a deliberate decision. `sp=` sets the policy for subdomains; left unset they inherit `p`, and a subdomain that sends through some other system will start failing when the parent tightens. RFC 9989 also adds `np=`, the policy for subdomains that do not exist at all, which is the tightest anti-spoofing control in the record because no legitimate mail can come from a name that does not exist. The record crawlcheck.io publishes today reads: ``` v=DMARC1; p=quarantine; sp=quarantine; pct=100; adkim=r; aspf=r; rua=mailto:dmarc@crawlcheck.io; fo=1 ``` `pct=100` applies the policy to everything and is harmless under either specification; on the next edit it can simply be removed. ## What will quarantine not fix? # DMARC governs the visible From domain only. A phishing message that uses a [lookalike domain](/glossary/lookalike-domain), or your brand name in the display name over a different address, passes DMARC because it is not claiming your domain. What tightening does is remove exact-domain forgery, the kind that most reliably defeats a reader, and remove your domain from the pool of spoofable senders. The neighbouring pieces, an [MTA-STS](/glossary/mta-sts) policy to stop transport downgrade and a [TLS-RPT](/glossary/tls-rpt) address to hear about it, are described on the [trust center](/trust); the single-character bounce that started this work is in [its own post](/blog/one-character-of-undeliverable-email). ## Checklist # - List every system that sends as your domain. - Read `Authentication-Results` on a real message from each, and get DKIM aligned on all of them, not just SPF. - Create the report mailbox and confirm it receives. - Publish `p=quarantine` (optionally with `t=y` first), read the reports, then tighten. - Set `sp=` and `np=` deliberately. A scan on [crawlcheck.io](/) reads the SPF and DMARC records as published and reports what policy is in force; it cannot see your senders, which is why step two is done by hand. ## The SPF lookup limit # SPF allows at most 10 mechanisms that need a DNS lookup (`include`, `a`, `mx`, `ptr`, `exists` and the `redirect` modifier), counted across every nested include. Past 10, the check fails with a permanent error, and DMARC then sees no SPF pass even for your own mail. Expand the whole include chain before tightening DMARC; adding one more sending service is the usual way a record crosses the limit. ([RFC 7208](https://www.rfc-editor.org/info/rfc7208/)) CrawlCheck’s dataset when this page loaded: 570,959 sites measured. Written by [Emmanuel Orta](/author) · VSNARY I build CrawlCheck and write everything published here, from measurements the verification layer took. Where a post got something wrong, the correction stays in the post. [Every post, and where I got it wrong →](/author) · [Follow on LinkedIn →](https://www.linkedin.com/in/emmanuelorta) Share this finding: [LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fcrawlcheck.io%2Fguides%2Fhow-to-move-dmarc-from-none-to-quarantine-without-losing-mail%3Futm_source%3Dlinkedin%26utm_medium%3Dsocial%26utm_campaign%3Dshare)[X](https://twitter.com/intent/tweet?url=https%3A%2F%2Fcrawlcheck.io%2Fguides%2Fhow-to-move-dmarc-from-none-to-quarantine-without-losing-mail%3Futm_source%3Dx%26utm_medium%3Dsocial%26utm_campaign%3Dshare&text=How%20to%20move%20DMARC%20from%20p%3Dnone%20to%20p%3Dquarantine%20without%20losing%20email)[Email](mailto:?subject=How%20to%20move%20DMARC%20from%20p%3Dnone%20to%20p%3Dquarantine%20without%20losing%20email&body=https%3A%2F%2Fcrawlcheck.io%2Fguides%2Fhow-to-move-dmarc-from-none-to-quarantine-without-losing-mail)[Follow Emmanuel on LinkedIn](https://www.linkedin.com/in/emmanuelorta) ## Questions this post answers ### What does p=none do? It asks receivers to deliver everything and send reports. It is a monitoring setting, not a protection; forged mail using the exact domain is delivered normally under p=none. ### What is DMARC alignment? The domain that passed SPF or DKIM must match the organisational domain in the visible From address. A sender can pass SPF or DKIM on its own domain and still fail DMARC because nothing aligned. ### Why does SPF alone fail when mail is forwarded? A forwarding server re-sends the message from its own IP, which is not in your SPF record, so SPF fails at the final receiver. A DKIM signature survives forwarding because it is computed over the message content, which is why aligned DKIM is what makes quarantine safe. ### Is the DMARC pct tag still valid? RFC 9989, published in May 2026, removes pct and replaces its testing use with the t= tag. Receivers may still honour pct during the transition, but new records should use t=y for testing. ### What does t=y do in a DMARC record? Under RFC 9989 it applies the next less strict policy: p=reject with t=y is enforced as quarantine, and p=quarantine with t=y as none. It is not an off switch like the old pct=0. ### Where do DMARC aggregate reports go? To the address in the record's rua tag, one XML report per receiver per day. The address must exist before the record names it; reports sent to a non-existent mailbox bounce and are lost. ### Should I go straight to p=reject? Only after quarantine has run long enough for the reports to show no legitimate source failing. Reject refuses the message outright, so a sender you forgot loses mail instead of having it filtered. ### What is the np= tag? Added by RFC 9989, np= sets the policy for subdomains that do not exist. Since no legitimate mail can come from a name that does not exist, it can usually be set to reject safely. ### How long does a DMARC change take to work? Receivers read the new record when their cached copy of the old one expires, which the TXT record's time-to-live setting controls. Lower that setting a day before the change if you want it to apply quickly. Aggregate reports usually arrive once a day, so allow at least 48 hours before judging the effect. ### How long should I stay at p=quarantine before p=reject? Until the reports show every legitimate sender passing for a full cycle of your mail, including monthly senders such as invoices and newsletters. For most domains that is several weeks. Move when the only failures left are messages you did not send. ## Related findings [2026-09-30 Five days of our alerts went to an address with no mailbox After a mail move, our alert address had no mailbox. For five days every alert said sent and none arrived. Why checks passed, and](/blog/five-days-of-our-alerts-went-to-an-address-with-no-mailbox)[2026-11-06 18 DMARC reports, 7 domains: every failure was a spoofer A DMARC reading across seven business domains: 18 reports, every failure from five outside addresses forging two brands, zero real](/blog/eighteen-dmarc-reports-every-failure-was-spoofing)[2026-09-02 How to read a CrawlCheck report: grade, fixes, evidence How to read a CrawlCheck report: what the grade and three stages mean, which tab to open first, and how to check or dispute a find](/guides/how-to-read-a-crawlcheck-report) The main product Found this on your own site? **We fix it for $100.** Scan free to see where you stand. The fix is one site, every finding implemented and re-measured, with a sealed before and after. [Fix it all — $100](https://buy.stripe.com/4gM00j7OjawhdLEdyR2VG0v) [What it covers](/pricing#fix) · [scan first, it is free](/) Free browser extension Check this on any site you open, **without leaving the page.** CrawlCheck for Chrome grades every site you visit A–F and shows, from your own browser, what search engines and AI assistants are actually served. Free, no account. [Add to Chrome — free](https://chromewebstore.google.com/detail/plgjengobhbdmbgckiiaaflgiepdbcpi?utm_source=crawlcheck.io&utm_medium=post) [What it does](/extension) Write for the library Measured something yourself? **Submit an article.** Send your finding, the data and the method. A person reads every submission and replies by email; accepted articles are published under your name with a link to your study. Free to submit. Articles are accepted on merit; no one can pay to have one published. [Submit an article](/submit-finding#send) [How it works](/submit-finding) · a reply within about a week Spotted a mistake or a bug on this page? [Report an error](/report-error?k=article). Confirmed errors are fixed and logged on the [corrections page](/corrections). ## Comments Comments are read before they appear. Nothing is published automatically, and no account is needed. An AI agent with a published signing key can post without the queue: it signs the request and the signature is checked against its own domain ([how](/api/comment/signed)). Those appear at once, marked *verified*. Send for review Writing about this? [Facts, live figures and marks](/press) — every number on that page is dated and traceable to a scan. [All guides](/guides) · [The dataset](/data) · [How the dataset works](/policy) We read your site the way crawlers do, from our own network, and every report says exactly what was asked for and what came back. [Who runs this](/author) · [Press kit](/press) · [Videos](/videos) · [Evidence for disputes](/evidence) · [YouTube](https://www.youtube.com/@crawlcheck) · [hello@crawlcheck.io](mailto:hello@crawlcheck.io) #### How we measure, and our own machine files CrawlCheck reads every site as itself. Its crawler-identity tests send named user-agent strings from its own network, never from an operator’s published range, so a site that verifies crawlers can tell them apart; every report says so. Results reflect what an anonymous non-browser client receives. [What we collect and how to opt out](/policy) · [how every report is timestamped](/proof) · [who runs this](/author) · [press kit](/press) · [hello@crawlcheck.io](mailto:hello@crawlcheck.io). **Second vantage:** the [browser extension](/extension) measures what crawlers are served from your own address and after your JavaScript runs — the half this server cannot see. From 1.0.0 it sends the domain you ask about, on one button, and anonymous usage counts you can switch off; from 1.0.3 a footer switch turns off the hostname lookup for the grade on the icon; from 1.0.4 a share switch, on by default, sends a signed Two-vantages reading (status, size, words and a hash per identity; never page text); from 1.0.16 chips send the hostnames of search results and AI citations while the chips and grade switches are on; from 1.0.17 a readings switch, on by default, sends one signed page reading a day (numbers, the hostname and the page address with the query removed; never in incognito; at most 40 a day) and only while the share and grade switches are also on. Earlier versions send nothing. **Our own machine layer for the agentic web:** [llms.txt](/llms.txt) · [llms-full.txt](/llms-full.txt) · [agents.md](/agents.md) · [SKILL.md](/SKILL.md) · [entitymap.json](/entitymap.json) · [media kit](/.well-known/media-kit.json) · [MCP server card](/.well-known/mcp/server-card.json) · [API catalog](/.well-known/api-catalog) · [OpenAPI](/openapi.json) · [robots.txt](/robots.txt) · [sitemap.xml](/sitemap.xml) · [security.txt](/.well-known/security.txt) — every one named in robots.txt, which is exactly what we check for on yours. **Pages:** [Keys](/keys) · [/](/) · [Listing check](/listing-check) · [SDK](/sdk) · [Conduct: ClaudeBot](/agents/conduct/claudebot) | **Registry:** [quickwebx.com](/registry/quickwebx.com) · [zervx.com](/registry/zervx.com) · [rankscale.ai](/registry/capabilities/rankscale.ai) | **Datasets:** [Technologies: Linkedin insight](/datasets/technologies/linkedin-insight) · [Technologies: Adobe fonts](/datasets/technologies/adobe-fonts) | **Blog:** [AI crawlers on Wix, Shopify, Squarespace, Webflow: 980 sites](/blog/what-ai-crawlers-get-from-wix-shopify-squarespace-webflow-sites) | **Glossary:** [entity map parity](/glossary/entity-map-parity) [Contact](/contact) · [Partnerships](/partnerships) · [Terms of service](/terms) · [Refund policy](/refunds) · [Trust center](/trust) · [Subprocessors](/subprocessors) · [Changelog](/changelog) · [Report an error](/report-error)
Same body as a file: /guides/how-to-move-dmarc-from-none-to-quarantine-without-losing-mail.md · as data: /feed.json · what we tell agents site-wide: /llms.txt, /agents.md. To see any other site this way, from your own browser, after its JavaScript runs: the browser extension.