Back to blog

Do TXT Record Lookups Help Identify a CDN? What Signals Does CDNChart Use?

CdnChart Technical TeamPublished on 2026-11-0911 min read2,047
Do TXT Record Lookups Help Identify a CDN? What Signals Does CDNChart Use?

When many people look up a domain's DNS, they see TXT records like the following:

provider-verification=xxxx
google-site-verification=xxxx
v=spf1 include:example.com ~all

If the name of a cloud provider or CDN vendor appears in one, it is easy to jump to the conclusion: 'This website is using that CDN.'

That conclusion can be useful as a reference, but it does not stand on its own.

TXT records are related to CDN identification, but they more often reflect domain verification, service configuration, or traces of past onboarding. They do not necessarily mean that current web traffic is passing through that CDN.To determine which CDN a website actually uses, you still need to combine the current CNAME, A/AAAA resolution results, response IPs, the ASN each IP belongs to, HTTP response headers, TLS certificates, and access results from different regions.

What exactly is a TXT record, and why does it appear in a domain?

According tothe definition of TXT records in RFC 1035, a TXT record is essentially one or more text strings associated with a domain name. DNS is only responsible for storing and returning these strings; what the content means depends on the service that publishes it and the use case.

Common uses of TXT records include:

  • Proving control of a domain;

  • Verifying a website or enterprise account;

  • Configuring an SPF email-sending policy;

  • Publishing DKIM or DMARC-related information;

  • Completing domain validation with a certificate authority;

  • Storing verification tokens for certain cloud services, SaaS platforms, or security products;

  • Recording administrator-defined notes.

Therefore, TXT is not a record type designed specifically for CDNs. When a browser accesses an HTTPS website, it usually needs to resolve A, AAAA, or CNAME records and then connect to the corresponding IP address. The browser does not automatically send requests to a CDN just because it sees a TXT record.

This is also why TXT records are better seen as 'configuration corroboration' than 'real-time traffic evidence.'

Why are TXT records often required when onboarding a CDN?

When onboarding a new domain, a CDN or cloud provider needs to confirm that the applicant actually owns the domain. Common approaches include:

  1. Requiring the user to add a TXT verification record;

  2. Requiring the user to add a specified CNAME record;

  3. Requiring the user to place a verification file in the website root directory;

  4. Completing verification through DNS resolution permissions or account authorization.

In some cases, the TXT record is only used for initial verification. After verification, the domain's actual traffic may enter the CDN through a CNAME, A record, or another routing method, but that TXT record may remain for a long time.

For example, a website may once have used a certain CDN and later migrated to another provider, but the administrator did not delete the old verification record. A TXT lookup will still show the old provider's string, even though the current web response already comes from the new CDN.

Certificate validation can also leave TXT records behind. For example, if a domain once used DNS-based validation to prove certificate issuance authority, the relevant record may be located at:

_acme-challenge.example.com

It only shows that a certificate validation once used a DNS challenge; it does not directly indicate which CDN the website currently uses.

Some DNS providers also use TXT records for domain verification, email authentication, and third-party service onboarding.In the DNS record type descriptionsyou can also see that TXT is not equivalent to a routing record for website access traffic.

When you see these TXT records, do not treat them directly as CDN evidence

The following types of TXT records are easily misjudged:

TXT record feature

More common actual use

Can it directly prove a CDN?

v=spf1

SPF email-sending policy

No

_dmarc.example.com

DMARC email security policy

No

selector._domainkey

DKIM email signing

No

google-site-verification

Search or site ownership verification

No

_acme-challenge

Certificate issuance or renewal validation

No

provider-verification

Domain control verification for cloud services or platforms

Only as a supporting clue

Customcdn=or vendor name

Administrator-defined notes or service verification

Requires other evidence for confirmation

For example, SPF itself is a mechanism that publishes email-sending policy through TXT records.The TXT lookup rules in the SPF standardcannot be used to prove that a website's web traffic passes through a certain CDN.

Therefore, after finding a TXT record containing a vendor name, the correct statement should be:

The domain has verification or configuration traces related to a certain provider, but the TXT record alone cannot confirm that current web requests still pass through that provider.

This is more accurate than simply saying 'this website uses such-and-such CDN.'

How do you specifically query TXT, CNAME, and the current resolved IP?

When troubleshooting, do not query only TXT. At a minimum, compare the TXT, CNAME, A, and AAAA records for the same hostname together.

On Linux, macOS, or an environment withdigthe required utility installed, you can use:

dig +noall +answer example.com TXT
dig +noall +answer www.example.com TXT

dig +noall +answer example.com CNAME
dig +noall +answer www.example.com CNAME

dig +noall +answer example.com A
dig +noall +answer example.com AAAA

On Windows, you can use:

nslookup -type=TXT example.com
nslookup -type=CNAME example.com
nslookup -type=A example.com
nslookup -type=AAAA example.com

Here is an easily overlooked issue:example.com,www.example.com,api.example.comandstatic.example.comare different DNS names; their TXT and CNAME records may be completely different.

When querying, record the following separately:

  • The full hostname queried;

  • The full content of the TXT record;

  • The final CNAME target;

  • The IPs returned for A and AAAA;

  • Query time;

  • The DNS resolver used;

  • Whether multiple results were returned.

If the domain itself exists but the TXT query returns an empty answer, this is not necessarily a query failure; it may simply mean that the hostname has no TXT record. You need to distinguish:

  • NOERRORbut no TXT answer: the domain exists, but that name has no TXT record;

  • NXDOMAIN: the queried name may not exist;

  • SERVFAIL or timeout returned: check the DNS service or network path;

  • Different resolvers return different results: this may be related to caching, TTL, split-horizon DNS, or routing policies.

You can also compare results from different public resolvers:

dig @1.1.1.1 +short TXT example.com
dig @8.8.8.8 +short TXT example.com

Different resolution results do not automatically mean DNS poisoning, nor do they necessarily mean the website uses multiple CDNs. You need to continue evaluating based on authoritative DNS, TTL, and other records.

Where does TXT fit when CDNChart identifies a CDN?

CDNChart's identification focus is not 'which record contains a vendor name,' but whether the current access path has a set of mutually supporting evidence.

InCDN identification methods and evidence combinations, the more valuable clues mainly include:

1. CNAME and DNS resolution chain

If the CNAME of a business domain points to a domain dedicated to a CDN vendor, this is a relatively direct configuration clue.

However, CNAMEs are not always publicly visible. Some platforms use DNS flattening, return A records, use Anycast addresses, or hide the real CNAME behind a proxy layer. Therefore, not finding a CNAME does not directly mean that no CDN is used.

2. Response IP and the ASN the IP belongs to

The IP actually returned when accessing the website can help determine what type of network the request enters. By combining the ASN the IP belongs to, network segment ownership, and results from different regions, you can observe whether edge network characteristics exist.

But IP ownership also has limits:

  • Multiple services may share the same cloud platform;

  • An ASN may host multiple products;

  • The IP registration location is not the same as the city of the node the user actually accesses;

  • A website may use multiple CDNs or cloud services at the same time;

  • The origin server may also be deployed in a large cloud provider's network.

So ASN is suitable for combining with other clues, not for drawing conclusions on its own.

3. HTTP response headers and cache characteristics

Some CDNs add specific response headers, cache status, request tracing IDs, or edge node information to responses.

You can use the following command to view response headers:

curl -sS -D - -o /dev/null https://www.example.com/

Focus on:

  • Server;

  • Via;

  • Age;

  • Cache-Control;

  • Vendor-specific tracing headers;

  • Whether the request is redirected;

  • Whether response headers vary by path or region.

A single response header should not be treated as absolute evidence. Response headers may be removed, overwritten, or spoofed, and they may also come from a reverse proxy at the origin rather than from the CDN itself.

4. TLS certificate and SNI information

TLS certificates can provide some clues about the current HTTPS entry point, such as the certificate subject, SAN, issuer, and the domains covered by the certificate.

However, certificates can also be shared. Many CDNs issue or host certificates for a large number of customers, and the same certificate authority may serve different platforms. Therefore, certificates are better used to confirm 'the service form of the current HTTPS entry point' and cannot be used to identify a CDN vendor based solely on the issuer.

5. Access results from multiple regions and carriers

If the same domain returns noticeably different IPs, ASNs, or response headers when accessed from different regions, DNS routing, edge nodes, or network paths may differ.

Such differences are important supporting information for judging CDNs. CDNChart's website speed test tool can observe response IPs, regions, carriers, HTTP status codes, DNS resolution time, connection time, and total time from different test nodes.

6. Whether the evidence supports one another

CDN identification is usually more reliable when:

CNAME指向某服务商
+ 多地区响应IP属于该服务商网络
+ HTTP响应存在相符特征
+ TLS和节点信息没有明显冲突

If there is only one TXT record and all other evidence points to another provider, the conclusion should not be written as 'confirmed.'

From this perspective, TXT records usually have lower priority: they can help explain how a domain once onboarded to a service, but they cannot replace checks of the current access path.

A more reliable CDN identification sequence

CDNChart recommends troubleshooting in the order of 'current traffic evidence first, configuration traces as supporting explanation.'

Step 1: Determine the business host to identify

First clarify what you want to query:

example.com
www.example.com
api.example.com
static.example.com

Do not apply the root domain's TXT records directly towwwanotherapisubdomain. Different subdomains may separately use the origin, a CDN, object storage, or third-party APIs.

Step 2: Query TXT, CNAME, A, and AAAA

Save the DNS results, paying particular attention to:

  • Whether the TXT record is for email authentication;

  • Whether there is a provider verification string;

  • Whether the CNAME points to a CDN domain;

  • Whether A and AAAA return different networks;

  • Whether the root domain and business subdomain use different configurations.

Step 3: Observe the actual HTTPS response

Use curl to request the real business URL and record the response IP and headers:

curl -sS -D headers.txt \
  -o /dev/null \
  -w "\nstatus=%{http_code}\nremote_ip=%{remote_ip}\n" \
  https://www.example.com/

If the website has login, API, or static asset domains, test these hosts separately. Testing only the homepage may miss the image, script, or API domains that actually use the CDN.

Step 4: Cross-check with a CDN detection tool

Submit the domain to theCDN detection tool, focusing on the public evidence the platform displays, such as CNAME, response IP, ASN, response headers, TLS, and nodes.

Here, look at whether the evidence is consistent, not just a single vendor name given by the tool. If the platform shows multiple candidate results, continue evaluating based on different subdomains, regions, and times.

Step 5: Use multi-region speed tests to observe current traffic

Usethe multi-region website speed test toolto compare response IPs, status codes, and latency across different regions and carriers.

If the TXT record shows provider A, but current response IPs and HTTP characteristics in all regions are closer to provider B, the following may be true:

  • The TXT record is an old verification record;

  • The website has migrated its CDN;

  • The root domain and business subdomain use different providers;

  • Static assets and dynamic APIs are connected to different CDNs;

  • The website uses multiple CDNs or a failover architecture.

In this case, do not force a single vendor choice. Instead, retain two conclusions: 'the currently observed service path' and 'the configuration traces shown by TXT.'

How should conclusions be written for different evidence combinations?

You can use the following wording according to evidence strength:

Observed result

More accurate conclusion

Only TXT contains a vendor name

Related verification or configuration traces found; current traffic cannot be confirmed to pass through that CDN

Both TXT and CNAME point to the same vendor

The service onboarding relationship is more credible, but current response IPs and HTTP characteristics still need to be checked

CNAME, IP/ASN, response headers, and TLS are all consistent

Multiple pieces of current access evidence support the CDN judgment

TXT points to A, while current response evidence points to B

There may be historical remnants, migration, multiple CDNs, or subdomain differences

No TXT, but current access evidence consistently points to a certain CDN

The absence of TXT does not affect identification based on current traffic evidence

No obvious CNAME or vendor response headers

This only means public fingerprints are insufficient; it cannot be concluded that there is no CDN

In the articleHow to judge an empty CDN detection result, the biggest misconception to avoid is interpreting 'no identification result' as 'the website does not use a CDN.' Private CDNs, self-built edge networks, proxy chains, multiple CDNs, and hidden response headers can all reduce the certainty of public identification.

FAQ

If Cloudflare or another vendor name appears in a TXT record, does that prove the website uses that CDN?

No. It may be a domain verification, certificate validation, historical onboarding, or configuration record for another cloud service. You need to continue checking the current CNAME, response IP, ASN, HTTP response headers, and TLS evidence.

If a website has no TXT records, does that mean it does not use a CDN?

No. Many CDN onboardings do not require public TXT records, or they use only CNAMEs, DNS routing, account authorization, and other verification methods. No TXT only means that no such text record was found.

Are SPF, DKIM, and DMARC records related to CDN identification?

Usually not directly. They are mainly used for email sending authorization, email signing, and anti-fraud policies. According tothe standard description of SPF records, SPF publishes email-sending policy through TXT and does not mean that web traffic passes through a certain CDN.

The root domain has a TXT record. Why does querying www not return one?

TXT records are bound to a specific DNS name.example.comThe root domain andwww.example.comwww are two different names; the root domain's TXT record is not automatically copied to all subdomains. When querying, you must use the actual business hostname.

If no CNAME is found, can a CDN still be identified?

Yes. DNS flattening, Anycast, proxy layers, and A-record routing can all make CNAMEs less visible. In this case, you can judge based on response IPs, ASN, HTTP response headers, TLS certificates, and multi-region access results.

Does CDNChart identify CDNs based only on TXT records?

It should not be understood that way. TXT is better used as a manually added configuration clue. CDNChart's judgment focuses on the combination of current DNS, network, HTTP, TLS, and multi-region access evidence; the specific conclusion also depends on the identification evidence and confidence actually displayed on the page.

If an old TXT record is still visible after a CDN migration, should it be deleted?

First confirm whether it is still used for certificate issuance, domain verification, or other business purposes. The old record may no longer serve a purpose, or it may still be relied on by a service. Do not delete or modify DNS records just because the record shows an old CDN name.

  • TXT records and CDN
  • DNS TXT records
  • CDN detection methods
  • CNAME lookup
  • CDN detection tools
  • CDN provider identification

Related posts