Skip to content
emailrules.today

Authentication history · 60 days observed

costco.com

Observed from 4 Aug 2026 to 3 Oct 2026. Nothing published in DNS has moved in that window.

Where it stands today

Live lookup, 3 Oct 2026. The same check /check/costco.com runs.

  • 3fine
  • 2context

  • Looks fine

    present, ending -all

    Hard fail. The strictest setting and the right one once you are confident every sender is listed.

    v=spf1 include:%{ir}.%{v}.%{d}.spf.has.pphosted.com -all

    See what this looks like →

  • Looks fine

    present with p=reject

    A policy that actually instructs receivers, which is more than most senders publish.

    v=DMARC1; p=reject; sp=reject; fo=1; rua=mailto:dmarc_rua@emaildefense.proofpoint.com; ruf=mailto:dmarc_ruf@emaildefense.proofpoint.com;

    From DMARC p=none is monitoring, not enforcementSee what this looks like →

  • Looks fine

    keys published on 4 selectors

    A key existing is not the same as working. Read a real received header and check the d= value matches your before you call this done.

    google._domainkey (Google Workspace), mandrill._domainkey (Mandrill), s2._domainkey (SendGrid), s1._domainkey (SendGrid)

    From DKIM passing is not DKIM alignedSee what this looks like →

    Part platform, part you

    The key is your platform's to publish and it has. Whether it signs the domain in your is yours to confirm, and cannot show it — send one campaign to yourself and look for =pass header.d=costco.com in the Authentication-Results header.

  • Context

    SendGrid signs your mail, and your cannot be read to confirm it

    2 of SendGrid's selectors carry live keys on this domain, so SendGrid is signing mail as you. Whether your authorises it is not answerable by reading . Your record uses SPF macros (Proofpoint), so the authorised senders are resolved per message from the connecting IP and are never published as a list. No checker can settle it from DNS, including this one — anyone who tells you this record does or does not list SendGrid is guessing.

    v=spf1 include:%{ir}.%{v}.%{d}.spf.has.pphosted.com -all

    From Gmail enforces authentication, PTR, TLS and a 0.30 percent spam rateSee what this looks like →

    Good to know — nothing to fix

    Send one real campaign through SendGrid and read the Authentication-Results header on what arrives. That header is the only place this question gets answered, because it is the receiver evaluating the macro against the real .

  • Context

    MX records present

    Where you receive mail says nothing about where you send it. Marketing sends usually leave through a different platform entirely.

    mx0a-001f9d01.pphosted.com, mx0b-001f9d01.pphosted.com

    See what this looks like →

What has moved

One entry per day a published record actually changed. Days we looked and found nothing different are counted, not listed.

  1. First observation — what was already published

    SPF published.

    v=spf1 include:%{ir}.%{v}.%{d}.spf.has.pphosted.com -all

    DMARC published.

    v=DMARC1; p=reject; sp=reject; fo=1; rua=mailto:dmarc_rua@emaildefense.proofpoint.com; ruf=mailto:dmarc_ruf@emaildefense.proofpoint.com;

    DKIM keys on selectors we probe.

    google._domainkey (Google Workspace), mandrill._domainkey (Mandrill), s1._domainkey (SendGrid), s2._domainkey (SendGrid)

    MX records present.

    mx0a-001f9d01.pphosted.com, mx0b-001f9d01.pphosted.com
Where this comes from. Public DNS, and nothing else. We read the same TXT and MX records any mail server reads before accepting a message, on the days someone looked. There is no scan, no login, no mail, and no score here — only what was published and the date we saw it. Gaps are days we did not get a clean answer from a resolver, and we would rather leave those blank than guess at them.
Run the full checkSee which rules are yours