DNS Propagation Explained — When Will My Records Be Live?

DNS propagation in one sentence#

When you publish a DNS record, the rest of the internet doesn't see it instantly — it takes minutes-to-hours for caches around the world to refresh and discover the new value.

Why caching exists#

Every DNS resolver caches responses. Without caching:

• Every email recipient's spam filter would re-query your SPF record on every message
• Your DNS host would get billions of queries per day
• DNS would be massively slow

Instead, resolvers cache for TTL (Time-To-Live) seconds before re-checking.

What TTL is#

TTL is published per-record. Sample:

Type:  TXT
Name:  yourdomain.com
Value: v=spf1 include:amazonses.com -all
TTL:   3600 (seconds = 1 hour)

This tells caches: "Remember this value for 3600 seconds. After that, refresh from authoritative DNS."

Typical TTLs:

• Short (300s = 5 minutes) — when expecting changes; faster propagation
• Default (3600s = 1 hour) — common balance
• Long (86400s = 24 hours) — for super-stable records

For initial publication, set TTL=300. After verified working, raise to 3600 or 86400 for cache-friendliness.

Propagation in practice#

When you publish a new record, the propagation timeline:

For high-stake records (SPF, DKIM), assume 1-4 hours is typical. 24 hours is the SLA most DNS hosts publish.

Verify in AcelleMail#

Open the sending-domain detail#

In AcelleMail's sidebar, Sending → Sending domains. The list shows every domain you've registered with status chips (Verified / Pending / Failed) and per-auth indicators:

Click into your domain row. The detail page surfaces exactly which DNS records to publish (TXT for SPF, CNAMEs for DKIM, TXT for DMARC) with copy-paste-ready values + current verification state per check:

After publishing, click Verify domain. AcelleMail's dig-check runs against multiple resolvers; if at least one returns the new value, AcelleMail considers it propagated.

If status stays Pending:

• Wait 30-60 minutes; re-click Verify
• After 4 hours, suspect a real DNS issue (record actually not published)
• After 24 hours, definitely investigate (something is wrong)

Diagnose with dig#

From your terminal:

# Check from your default DNS resolver
dig TXT yourdomain.com +short

# Check from Cloudflare's 1.1.1.1
dig TXT yourdomain.com @1.1.1.1 +short

# Check from Google's 8.8.8.8
dig TXT yourdomain.com @8.8.8.8 +short

# Check from your local ISP's resolver (if you can identify it)
dig TXT yourdomain.com @your-isp-dns +short

If all 4 return the new value: fully propagated. AcelleMail's pending state should resolve immediately.

If some return new + some return old: partial propagation — keep waiting.

If all return old (or nothing): the record isn't actually published. Check your DNS host's admin panel.

What controls propagation speed#

If your previous record had TTL=86400 (24 hours), THAT determines max wait time for the change — even if your new record has TTL=300, old caches will hold the previous value for up to 24h.

Always lower TTL BEFORE making a change, then change, then optionally re-raise:

Day 0: Lower TTL on existing record to 300 (5 min)
Day 1: Wait 24 hours for that low TTL to fully propagate
Day 2: Publish your change → propagates in ~5 minutes
Day 3: Optionally raise TTL back to 3600

For one-off setup (first-time SPF/DKIM/DMARC), skip the pre-stage; publish directly + accept the wait.

Common propagation issues#

What you CAN'T do#

When to consider longer waits#

DNS propagation can occasionally take longer than expected. Consider:

• Region-specific delays — sometimes APAC propagation lags US by 6-12 hours
• Recently-changed authoritative DNS — if you recently switched DNS hosts, propagation takes longer
• Vendor-specific delays — some authoritative DNS hosts have slower-than-typical refresh rates

If after 48 hours something is still Pending, contact your DNS host's support.

Day -7: Lower TTL on the existing record to 300 (5 minutes)
        This change takes 24h+ to propagate to all caches.

Day -6 to -1: Wait. All caches eventually pick up TTL=300.

Day 0: Publish the actual change.
       Propagation: ~5 minutes globally (because all caches now hold TTL=300).

Day +1: Optionally re-raise TTL to 3600 for cache-friendliness.

#!/bin/bash
# Check propagation across multiple public resolvers
resolvers=("1.1.1.1" "8.8.8.8" "9.9.9.9" "208.67.222.222")
record="yourdomain.com"
expected="v=spf1 include:amazonses.com -all"

for r in ${resolvers[@]}; do
  actual=$(dig TXT $record @$r +short | head -1 | sed 's/"//g')
  if [ "$actual" == "$expected" ]; then
    echo "✓ $r: matches"
  else
    echo "✗ $r: still propagating (got: $actual)"
  fi
done

Related articles#

• How to set up SPF, DKIM, and DMARC records
• Complete DNS setup for email sending
• DKIM key rotation playbook
• Sender reputation monitoring