How to Troubleshoot DNS Changes That Have Not Propagated
DNS troubleshootingDNS propagationnameserversTTLDNSSEC

How to Troubleshoot DNS Changes That Have Not Propagated

SSiteHost Editorial Team
2026-11-15
7 min read

Diagnose DNS changes that are not working by checking nameservers, authoritative records, TTLs, caches, conflicts, and DNSSEC step by step.

DNS changes that have not propagated are usually diagnosable, but the right fix depends on where the answer is coming from. This checklist shows how to distinguish caching from an incorrect record, an inactive nameserver change, a conflicting entry, or a DNSSEC problem—without repeatedly editing settings and making the situation harder to read.

Overview

DNS translates a hostname such as www.example.com into information that other systems can use, including an IP address, mail server, or verification value. When you update a DNS record, different resolvers may continue returning the previous answer until the record’s time to live (TTL) expires. That delay is commonly called propagation, although it is more accurately a combination of caching and resolver behavior.

Before changing anything, identify the exact symptom and the record involved. Is the main domain failing, or only a subdomain? Is the website wrong while email works? Does a public DNS checker show the new value while your laptop shows the old one? These details determine whether you should inspect the authoritative zone, the domain’s nameservers, a local cache, or the application and hosting configuration behind the record.

A simple DNS troubleshooting decision tree

  1. Does the authoritative nameserver return the new record? If no, check the DNS zone, record name, value, and provider where the active zone is hosted.
  2. Do public resolvers return different answers? If yes, the change may still be cached according to the previous TTL. Compare several resolvers and allow time for caches to refresh.
  3. Do only local devices show the old answer? Clear or bypass the local DNS cache, test another network, and inspect browser or operating-system DNS settings.
  4. Is the domain returning an error rather than an old value? Check nameserver delegation, DNSSEC status, missing records, and authoritative server availability.
  5. Does DNS resolve correctly but the website or email still fail? Move to hosting, firewall, TLS, web-server, mail-server, or application configuration rather than continuing to edit DNS.

Checklist by scenario

1. A website points to the old host

  1. Confirm the intended hostname, such as the apex domain, www, or a separate subdomain. These may have different records.
  2. Check the authoritative answer with commands such as dig example.com A and dig www.example.com A. For IPv6, also check AAAA.
  3. Query the domain’s authoritative nameserver directly. First find it with dig example.com NS, then query one of the returned servers: dig @ns1.example-dns.com example.com A.
  4. Compare the authoritative result with public resolvers, for example dig @1.1.1.1 example.com A and dig @8.8.8.8 example.com A. Resolver addresses vary by environment, so treat these as test examples rather than a complete list.
  5. Check whether an old A record, AAAA record, or CNAME is still active. A device using IPv6 may reach the address in an AAAA record even when the A record is correct.
  6. Verify that the new hosting server accepts the requested hostname and that its firewall, web server, redirects, and SSL certificate are configured for it.

2. A new subdomain does not work

Confirm the record was created in the active DNS zone and that the name was entered as intended. Control panels may label the name field as Host, Name, or Subdomain. Some interfaces automatically append the domain, so entering the full name can accidentally create a duplicated hostname. Query the exact subdomain directly:

dig app.example.com A
dig app.example.com CNAME

If the response is NXDOMAIN, the name does not exist from the resolver’s perspective. If it returns a value but the service fails, inspect the destination server and application. If the subdomain is delegated to another DNS provider, check for an NS record at that label and troubleshoot the delegated zone separately.

3. Email stopped after a DNS change

Review the MX records first, including their priority values and destination hostnames. Then check the destination hostnames’ A or AAAA records. Do not replace email records with website records simply because the website is working. Also inspect TXT records used for mail authentication, such as SPF, DKIM, and DMARC, when the symptom involves rejected or quarantined messages.

Use dig example.com MX and query each mail destination separately. A DNS answer can be correct while the mail provider is unavailable, the mailbox is misconfigured, or an authentication policy is too restrictive.

4. A domain shows DNS errors after changing nameservers

Nameserver changes are different from ordinary record edits. Check the registrar’s delegation for the domain with:

dig example.com NS
dig +trace example.com

Make sure the nameservers shown at the registrar are the ones hosting the intended zone. Then confirm those authoritative servers respond with the required records. A frequent cause of DNS nameserver problems is editing a zone at one provider while the domain is delegated to another.

Record type and symptom comparison

Record or settingCommon symptomWhat to check
AIPv4 website points to the wrong serverAddress, hostname, TTL, and duplicate records
AAAASome users cannot reach the site or reach an old hostIPv6 address and whether the server supports IPv6
CNAMESubdomain resolves to an unexpected destinationTarget name, conflicting records, and alias restrictions
MXMail delivery fails or goes to the wrong providerPriorities and destination host records
TXTVerification or email authentication failsExact value, quotation handling, and multiple TXT entries
NSZone changes appear to have no effectRegistrar delegation and authoritative zone provider
DNSSECValidation errors or widespread resolution failureDelegation signer data, keys, and provider status

What to double-check

TTL and cache behavior

TTL is a caching instruction expressed in seconds. A resolver may retain an old answer until its cached TTL expires, though actual behavior can vary by resolver and infrastructure. Lowering the TTL immediately before a change does not remove values already cached under the previous TTL. Plan DNS migrations in advance when possible, and avoid repeatedly changing a record while waiting for results; each edit makes diagnosis less certain.

Authoritative versus recursive answers

The authoritative nameserver is the source for the zone. A recursive resolver answers clients and may return a cached result. Always compare both. If the authoritative server is wrong, waiting will not solve the problem. If the authoritative server is correct but recursive resolvers differ, caching is a more likely explanation.

Conflicting records

Look for more than one record of the same type, an unexpected wildcard such as *.example.com, or an old record at a different hostname. A CNAME generally should not be combined with other data at the same name. Also check whether the control panel displays flattened, proxied, or synthesized records that differ from the value you entered.

Local caches and test conditions

Test from a second network or device before clearing anything. If only one environment is affected, flush the operating-system DNS cache, restart the relevant browser or application, and inspect VPN, corporate DNS, security software, and router settings. Browsers and applications can maintain their own short-lived caches.

DNSSEC status

DNSSEC can cause validation failures when the parent-zone delegation contains stale signing information after a DNS provider change. If some validating resolvers report a DNSSEC-related error while others appear to work, inspect the domain’s DS records at the registrar and the DNSSEC keys at the authoritative provider. Do not remove DNSSEC records casually; coordinate the disable, rollover, or migration sequence with the provider that manages the zone.

Common mistakes

  • Editing the wrong provider: The registrar, DNS host, and web host may be separate services. Identify which nameservers are delegated before editing records.
  • Checking only the browser: A browser test does not show which DNS answer was used. Use direct DNS queries and compare authoritative and recursive results.
  • Ignoring the apex and www difference: example.com and www.example.com can resolve independently.
  • Forgetting IPv6: A correct IPv4 record cannot compensate for an incorrect IPv6 record when clients prefer IPv6.
  • Deleting records to force propagation: Unnecessary edits can interrupt websites, email, verification, and certificates. First record the current values and change only the identified problem.
  • Assuming DNS is the whole failure: Resolution proves only that a destination was found. It does not prove that the destination is serving the right site or accepting traffic.

When to revisit

Use this checklist whenever you change hosting, launch a subdomain, move email, verify a service, renew or transfer a domain, or prepare a DNS migration. Revisit it before seasonal planning cycles and after changes to a registrar, DNS control panel, hosting platform, resolver setup, or DNSSEC workflow. Command syntax and control-panel labels can change, so review your internal runbook at least annually.

For a controlled change, save the old records, intended new records, TTLs, authoritative nameservers, and test results. After editing, verify the authoritative answer first, then check several recursive resolvers and finally test the actual website or mail flow. If the issue remains unresolved, provide your hosting or DNS support team with the hostname, record type, timestamps, authoritative results, resolver results, and any DNSSEC error. That evidence is more useful than making another untracked change.

Once the domain is stable, document who controls registration, nameservers, DNS records, hosting, and renewal notifications. This creates a practical baseline for future domain and hosting changes and reduces the chance that a routine update becomes an avoidable outage.

Related Topics

#DNS troubleshooting#DNS propagation#nameservers#TTL#DNSSEC
S

SiteHost Editorial Team

Technical Editor

Senior editor and content strategist. Writing about technology, design, and the future of digital media. Follow along for deep dives into the industry's moving parts.