DNS_PROBE_FINISHED_NXDOMAIN: Causes and How to Fix It

· 5 min read · 10 views · Getwebup

You open the site, Chrome throws "This site can't be reached - DNS_PROBE_FINISHED_NXDOMAIN", and now you're stuck guessing whether it's your domain, your DNS, or just your laptop being weird. Here's how to actually narrow it down instead of poking at settings at random.

What NXDOMAIN actually means

NXDOMAIN is a real DNS response code, not a Chrome invention. It means the resolver asked "what's the IP for this hostname?" and got back "this name doesn't exist" - not a timeout, not a server error, an explicit no. Chrome's DNS_PROBE_FINISHED_NXDOMAIN is just its way of surfacing that response after it double-checked your connection is otherwise fine.

That distinction matters because it rules out a whole category of fixes. If the domain truly doesn't resolve, restarting your router won't help. If it resolves fine everywhere except your machine, changing nameservers won't help either. You need to know which side of that line you're on before you touch anything.

Step 1: Check if it's you or everyone

Open a terminal (or Command Prompt on Windows) and run:

nslookup yourdomain.com

Or, better, use a third-party lookup that isn't affected by your local cache - dig yourdomain.com from a VPS, or an online tool from a different network entirely (your phone on mobile data works too).

  • Resolves fine elsewhere, fails only on your machine - it's local: browser cache, OS DNS cache, or your ISP's resolver having a bad day. Skip to the local fixes below.
  • Fails everywhere - it's the domain's DNS setup, not your device. Skip to the domain-side causes.

Common cause #1: Nameservers point somewhere with no records

This is the single most common reason for a domain-wide NXDOMAIN. It usually happens after:

  • You changed nameservers (e.g. moved from your registrar's default NS to Cloudflare, or to Getwebup's) but never added the A/CNAME records at the new provider.
  • You migrated hosting and the old DNS zone got deleted before the new one was fully populated.
  • A typo in one of the nameserver hostnames at the registrar.

Check which nameservers are actually live right now:

dig NS yourdomain.com +short

Then query that same set of nameservers directly for your A record:

dig @ns1.example-dns.com yourdomain.com A

If that comes back empty, the nameservers are correct but the zone at that provider has no A record for the root domain - go create one. If the NS records themselves look wrong (pointing to servers you don't recognize), fix that at your domain registrar first; nothing else will work until the correct nameservers are live.

Common cause #2: www vs. root domain mismatch

NXDOMAIN on www.yourdomain.com while yourdomain.com works fine (or vice versa) usually means one of the two records is simply missing. Root domains need an A record (or ALIAS/ANAME if your DNS provider supports it); www is almost always a CNAME pointing back to the root, or its own A record. Open your DNS zone editor and confirm both exist - it's an easy one to miss when you're moving fast during a migration.

Common cause #3: Newly registered or renewed domain

If you just registered the domain, it can take anywhere from a few minutes to a few hours for the registry to propagate the initial NS delegation - NXDOMAIN during that window is normal, not a misconfiguration. Same goes right after a domain expires and gets renewed; give it time before troubleshooting further. You can watch propagation status with a tool like whatsmydns.net, checking multiple regions at once.

Common cause #4: Typo in the domain itself

Obvious, but worth ruling out early - a wrong TLD, a transposed letter, or an old bookmark pointing at a domain you no longer own will all produce a clean NXDOMAIN, because as far as DNS is concerned, that name genuinely doesn't exist. Triple-check the exact string before assuming it's a config issue.

Local-side fixes (when it's just your machine)

If the lookup fails only for you, work through these in order:

  1. Flush Chrome's internal DNS cache - go to chrome://net-internals/#dns and click "Clear host cache".
  2. Flush the OS DNS cache. On Windows: ipconfig /flushdns. On macOS: sudo dscacheutil -flushcache; sudo killall -HUP mDNSResponder. On Linux (systemd-resolved): sudo resolvectl flush-caches.
  3. Switch your DNS resolver. Some ISP resolvers are slow to pick up changes or occasionally misbehave. Try 1.1.1.1 (Cloudflare) or 8.8.8.8 (Google) in your network adapter settings and retest.
  4. Restart the router if it caches DNS locally - many consumer routers do, and a reboot clears it.
  5. Disable a misbehaving VPN or browser extension temporarily - some DNS-over-HTTPS extensions and ad blockers rewrite or intercept lookups in ways that produce exactly this error.

Quick diagnosis table

SymptomLikely causeWhere to fix it
Fails on every device/networkMissing A record or bad nameserversDNS zone editor / registrar
Fails only on your machineLocal DNS cache or resolverBrowser/OS/router
www fails, root works (or reverse)Missing CNAME or A record for one of the twoDNS zone editor
Started right after registering/renewingRegistry propagation delayWait, then recheck
Started right after a migrationOld zone removed before new one was liveDNS zone editor at new host

Prevention

Most NXDOMAIN incidents we see at Getwebup trace back to one habit: changing nameservers before the destination zone is fully built out. If you're migrating a domain, add every record at the new DNS provider first, confirm they resolve with dig against that provider's nameservers directly, and only then update the NS records at your registrar. That way there's never a window where the domain points somewhere empty. It's also worth keeping TTLs on critical records (A, MX) around 300-3600 seconds rather than the default 24 hours, so that if something does go wrong, a fix propagates in minutes instead of a full day.

Questions people actually ask