Global DNS Propagation Check
Real dig queries from six countries, so 'has it propagated' gets a real answer.
The domain to resolve
How this check works
After a DNS change, the question that matters is what resolvers in other places actually say — not what one resolver near you says six times. Each run here asks the Globalping probe network to execute a real dig from machines on different continents, each against the resolver that machine genuinely uses, the way a visitor in that country would resolve your domain. The answers come back side by side with the resolver and timing for each. When they all agree, the change has propagated as far as these vantage points can see; when they disagree, you are watching caches expire in real time. The TTL in the result is the key number — it is the longest any resolver is allowed to keep serving the old answer, so it is also the honest answer to 'how much longer'.
What the results mean
- Distinct answers
- How many different responses the locations gave. One is what settled DNS looks like. Several, shortly after a change, is propagation under way.
- TTL (time to live)
- How many seconds a resolver may cache the answer. A change made with a TTL of 86400 can lawfully take a full day to be seen everywhere — that delay was set before the change, not caused by it.
- Resolver
- The DNS server the probe's network uses — often a public one like 8.8.8.8, sometimes an ISP's own. Each answer is what that resolver is currently serving to its users.
- No answer / NXDOMAIN
- The resolver says the record does not exist. Right after creating a record, this means the authoritative servers had not been updated when that resolver last asked — or the record name has a typo.
Common problems and fixes
- I changed the record hours ago and one region still shows the old value
- Look at the TTL. If the old record carried 86400, resolvers that cached it can serve it for 24 hours and are behaving correctly. Nothing accelerates a cache you do not control — the fix is prevention: drop the TTL to 300 a day before the next planned change.
- Every location shows the old value
- Then the change likely never reached the authoritative servers. Check you edited the right zone — the classic misses are editing at the registrar while the domain's NS records point at a different DNS host, or a forgotten trailing dot turning www into www.example.com.example.com.
- Regions disagree and it never settles
- Stable, geographic disagreement is usually geo-DNS: CDNs answer each region with a nearby endpoint deliberately. If you did not expect it, look at whether the record is a CNAME into a CDN — that is where the behaviour comes from.
- My own browser still opens the old site after everything here agrees
- Your machine has caches below DNS: the OS resolver cache and the browser's own. Flush them or wait them out — and if the site was moved, the browser may also be holding a redirect cache, which a private window sidesteps.
Frequently asked questions
How long does DNS propagation actually take?
Exactly as long as the old record's TTL, plus the moments it takes authoritative servers to sync — usually seconds. 'Up to 48 hours' is folklore from an era of long default TTLs. A record with TTL 300 is globally current within five minutes of the change reaching the authoritative servers; one cached with TTL 86400 takes up to a day, by design.
Can I speed up propagation after making a change?
For resolvers you do not run — no. Each holds its cached copy until the TTL it was given expires; that is the contract that makes DNS scale. What you can do is plan: lower the TTL well before a migration, make the change, confirm it here, then raise the TTL again.
Why do these results differ from my local dig?
Your dig asks your resolver, which has its own cache with its own age. A probe in another country asks a different resolver with a different cache. During the propagation window both are honestly reporting what they hold — which is why checking only from your own machine answers a narrower question than it appears to.
Is six locations enough to say it has propagated?
For the practical question — is the new answer being served across continents by unrelated resolvers — yes, agreement across six independent vantage points is strong evidence. It is not a census of every resolver on Earth, and the stragglers it cannot see are bounded by the same TTL arithmetic as everyone else.