DNS problems often look similar from the browser but have different causes. This guide compares practical DNS troubleshooting tools, shows how to check delegation and authoritative answers, and provides a repeatable workflow for investigating propagation delays, CNAME conflicts, incorrect nameservers, and DNSSEC failures.
Overview
DNS troubleshooting is most reliable when you separate the path into stages: parent-zone delegation, authoritative nameservers, recursive resolvers, and the final application endpoint. A browser-based DNS propagation checker can reveal that several locations receive different answers, but command-line tools usually provide the detail needed to explain why.
The main tools have different strengths:
| Tool | Best use | Resolver selection | DNSSEC visibility | Scripting |
|---|---|---|---|---|
dig | Detailed record queries, delegation checks, and repeatable diagnostics | Explicit resolver with @server | Strong; request DNSSEC records with +dnssec | Excellent |
nslookup | Quick interactive or platform-available lookups | Configurable, though syntax varies by use | Limited compared with dig | Usable |
host | Short, readable answers for common records | Usually supports a named server argument | Basic | Good for simple checks |
| Online DNS checker | Comparing answers from multiple public locations | Selected by the service | Varies by provider | Usually limited |
| DNS provider dashboard | Reviewing configured records, zones, and DNSSEC settings | Shows the provider's authoritative configuration | Often includes status and key settings | Depends on API support |
Use online developer tools for fast external comparison, but do not treat a checker map as proof that every user has the same result. Recursive resolvers cache answers independently, and a result can also differ because of record type, queried hostname, or an authoritative configuration issue.
What to track
1. Authoritative nameservers and delegation
Start with the nameservers published for the domain at its parent zone. The command below asks a recursive resolver to return the delegation:
dig NS example.com
Then query each listed authoritative server directly:
dig @ns1.example-dns.com example.com SOA
dig @ns1.example-dns.com www.example.com A
dig @ns2.example-dns.com www.example.com A
Replace the names and record types with your own values. If the authoritative servers disagree, investigate the zone configuration or provider synchronization before focusing on propagation. If the parent delegates to nameservers that are not serving the expected zone, the problem is higher in the DNS chain.
2. Record values and aliases
Track the records that matter to the application rather than only the apex domain. A web deployment may depend on an apex A or AAAA record, a www CNAME, a staging hostname, and one or more verification records. Email and API integrations may add MX, TXT, or service-specific aliases.
dig example.com A
dig example.com AAAA
dig www.example.com CNAME
dig example.com MX
dig example.com TXT
Record conflicts are often naming mistakes. A CNAME is an alias and generally cannot coexist with other data at the same owner name. Check whether the requested hostname is actually the alias you intended, whether a trailing dot appears in provider interfaces, and whether an old record remains at a nearby hostname.
3. TTL and recursive-resolver answers
The time-to-live, or TTL, is the remaining cache lifetime shown in a DNS response. It is not a guaranteed global propagation timer. Compare authoritative answers with several recursive resolvers:
dig @1.1.1.1 www.example.com A
dig @8.8.8.8 www.example.com A
dig @9.9.9.9 www.example.com A
Use resolver addresses appropriate to your environment and verify them rather than copying examples blindly. If recursive answers differ from the authoritative answer, note the returned TTL and the time of each test. A resolver may still be serving a cached response, while another may have refreshed sooner.
4. DNSSEC status and the validation path
When DNSSEC is enabled, check both the signed zone data and the delegation chain. Request DNSSEC-related records with:
dig example.com DNSKEY +dnssec
dig example.com SOA +dnssec
dig DS example.com
A DNSSEC failure can result from stale or incorrect delegation data, a missing or mismatched DS record, expired signing material, or an incomplete change at the DNS provider. The exact failure is easier to isolate when you compare the parent delegation, authoritative response, and validating recursive response. Avoid disabling validation as a casual fix; use it only as a controlled diagnostic comparison and restore the intended security configuration.
5. The final endpoint
DNS only maps a name to an address or alias. After resolving the name, confirm that the returned endpoint serves the expected application. For HTTP services, compare the address with the configured CDN, load balancer, or host and then test the request:
curl -I https://www.example.com
An HTTP error, certificate mismatch, redirect loop, or virtual-host configuration problem can look like a DNS outage. Keep DNS evidence separate from application and TLS evidence.
Cadence and checkpoints
DNS does not need constant manual inspection, but a lightweight record of expected values makes changes easier to review. Run a baseline check when a domain or subdomain is created, and repeat it after nameserver, CDN, hosting, email, or DNSSEC changes.
- Monthly or quarterly: check critical hostnames, authoritative nameservers, important record types, and the expected application endpoint.
- Before a migration: record current answers, TTLs, delegation, DNSSEC state, and the provider responsible for each zone.
- During a rollout: query authoritative servers and selected recursive resolvers at consistent intervals, recording timestamps and results.
- After a change: confirm that the intended records are visible, old endpoints are no longer returned where they should not be, and the application responds correctly.
- When alerts appear: compare the failing hostname and record type from the report with direct authoritative and recursive queries.
For repeatable checks, save command output with a timestamp and environment details. A small shell script using dig can compare expected nameservers or addresses without requiring a full monitoring system. Do not store secrets in diagnostic scripts, and treat public DNS output as potentially sensitive operational information.
How to interpret changes
Only one recursive resolver differs: this often points to caching or resolver-specific behavior. Compare TTLs and wait for the relevant cache lifetime before declaring the change complete.
Every recursive resolver differs from the authoritative server: verify that you queried the correct authoritative nameserver and that the parent delegation points to it. An incorrect nameserver at the registrar or parent zone can make a valid provider dashboard configuration irrelevant to public resolution.
Authoritative servers disagree: investigate zone replication, multiple DNS providers, or an unintended split configuration. Treat inconsistent authoritative answers as a configuration issue rather than ordinary propagation.
A hostname returns an unexpected CNAME: inspect aliases at each level and check whether a CDN, hosting platform, or redirect service owns the target. Follow the chain with dig and verify the final address.
DNSSEC-aware resolvers fail while non-validating tests appear to work: inspect DS and DNSKEY records, key rollover state, signatures, and the authoritative provider's DNSSEC status. The distinction is important because a response that resolves in a non-validating test may still fail validation.
DNS resolves but the site is unavailable: move to HTTP, TLS, firewall, load balancer, or origin checks. DNS troubleshooting should establish what address was returned; it cannot confirm that the service behind that address is healthy.
When to revisit
Revisit this workflow on a monthly or quarterly schedule for critical domains, and immediately after any change to hosting, CDN routing, registrar nameservers, email providers, staging environments, or DNSSEC. Also repeat the checks when users report intermittent failures, when only one region appears affected, or when a deployment changes the public hostname.
Keep a short baseline containing the domain, expected nameservers, important records, resolver test results, and the date of the last successful application check. During the next review, compare current output with that baseline rather than relying on memory. Update the commands and resolver list if your operating system, provider interface, or internal monitoring process changes.
A practical incident sequence is: query the affected name and record type, inspect authoritative answers, compare two or more recursive resolvers, check delegation and DNSSEC when relevant, then test the final HTTP or application endpoint. This order prevents a browser symptom from sending the investigation straight to the wrong layer.