How ipaddress.si works

Like every website, our server receives a few things with each request:

• Your IP address — the public address your connection uses. Behind a home router, a mobile network or a VPN this is the router’s, carrier’s or VPN’s address, not your device’s.
• Request headers your browser sends — user agent, preferred languages, and privacy signals such as Do Not Track and Global Privacy Control.
• The page you asked for, including anything you typed into a lookup.

Your browser also knows things it never sends us — your screen size, its time zone, the HTTP protocol it negotiated. Where we show those, they are computed in your browser and stay there.

• Our own copies of public datasets, downloaded and compiled on our server every 24 hours: the global BGP routing table and AS names (APNIC), address delegations from the five regional internet registries, IP geolocation (DB-IP Lite), place names and time zones (GeoNames), the Tor exit list and the IEEE MAC vendor registry.
• Aggregate page counts: the path of each page view (with lookup values removed, e.g. /ip/[query]) and a daily counter. No IP, no cookie, no identifier. Disabled when your browser sends Do Not Track or Global Privacy Control.
• Data source statistics: how many requests each source answered and how long it took, per hour — needed to monitor reliability. No lookup values are kept.
• Short-lived caches of results in memory (minutes to hours, never written to disk), so the same question isn’t asked twice.
• Messages you send through the contact form, until they are answered (deleted after at most 12 months).

• We do not log visitor IP addresses in our application. Rate limiting works on keyed hashes held in memory, rotated daily.
• We do not keep a history of the IPs, domains or ASNs you look up.
• We do not set tracking cookies. The only cookies are a language preference you choose, and — only if you accept — those set by Google Analytics or AdSense when the site operator has enabled them.
• We do not sell or share any visitor data.

Infrastructure in front of the application (the reverse proxy and hosting provider) may keep standard access logs for security and operations, with short retention.

1. Your input is validated strictly. Only IP addresses, ASNs and syntactically valid public domain names are accepted; private, internal and reserved names never leave our server.
2. Network, routing, registry, location and Tor data come from our own datasets on our server. No third-party IP intelligence service is asked about your address.
3. Two things are inherently live and are queried at the moment you ask: DNS (forward and reverse lookups, sent by our own DNS client to recursive resolvers) and registration records (RDAP or WHOIS, asked directly of the registry responsible for the address, AS number or domain).
4. Every field shows where it came from. Values that are inferred rather than recorded — such as the network type guessed from an operator’s name — are labelled as a guess.
5. If a source fails or is slow, the page shows what is available and marks the rest as not available — never a made-up value.

Commercial data providers can be connected by the site operator as an optional extra; they are switched off by default and listed below with their current status.