What we checkDNS hygieneDN-01
Name server redundancy
At least two name servers, on at least two networks.
- Area
- DNS hygiene
- Check
- DN-01
- Included
- Free in every scan
What we check
We read your NS records, resolve each one, and check how many distinct networks they sit on.
Why it matters
If your DNS goes down, your domain does not slow down — it disappears. The website, the email, everything that resolves your name.
Two name servers in the same rack are one failure away from that. The redundancy has to be somewhere else.
The failure is also total rather than partial. An overloaded web server serves some requests slowly; a name server that cannot be reached means resolvers have no address to try at all, and the domain simply does not exist for the duration.
Recovery is slow too: resolvers cache the failure, so service comes back gradually rather than the moment you fix it.
How to fix it
- 1
Use a provider with name servers on separate networks — most managed DNS is like this by default.
- 2
If you run your own, put the secondary somewhere genuinely independent.
Common questions
- My provider gives me four name servers on one prefix.
- That is common and it is what this check is about: four addresses in one network share one failure. Ask your provider about their anycast footprint, which may mean more diversity than the addresses suggest.
- Do I need a second DNS provider?
- For most sites, no. For anything where an hour of downtime is expensive, a second provider is the standard answer.
- Why did a name server show as unresolved?
- We could not get an address for it. That is either a stale NS record for a server that no longer exists, or a real resolution failure — both worth checking.
See how your domain does on this check.
All 34 checks, a grade, and the exact fix for anything that isn’t right — in about twenty seconds, no signup.