Why IP Geolocation Can Be Wrong After an IPv4 Transfer

An IPv4 transfer changes registry data, but GeoIP providers may continue to show the old country. Learn which records matter and how to request corrections.

IP geolocation after an IPv4 transfer does not update everywhere at the same time. The Regional Internet Registry may show the new resource holder while a website, advertising platform or streaming service still assigns the block to its previous country. This can happen because registry records, routing data, operator-published geofeeds and commercial GeoIP databases are separate systems.

To correct an outdated location, the recipient should verify the RIR record, publish an accurate geofeed, check the live BGP announcement and submit evidence to the databases used by affected services. No single change can force every provider to update, and no broker can guarantee one universal correction date.

Why IP Geolocation After an IPv4 Transfer Can Be Wrong

An IPv4 address transfer changes the registration of an eligible prefix from a source organisation to a recipient organisation. It does not automatically rewrite every third-party database that has previously associated the prefix with a country, city, company or connection type.

GeoIP providers build their own datasets from multiple signals. These may include operator submissions, registry records, routing observations, network measurements, customer feedback and historical evidence. Providers use different sources, confidence rules and release schedules, so two current databases can return different answers for the same IP address.

A transfer can also be followed by a genuine change in deployment. A block previously used in one region may be deployed in another country or operated by a different network. Until providers obtain and process evidence of that change, historical location data can remain visible. However, a change of registered holder alone does not necessarily mean that the block’s geographic location has changed.

Why Does WHOIS Show the Right Country but the Website Still Shows the Wrong One?

RIR databases primarily document Internet number-resource registration and contacts. A country: value may describe an administrative or registration context rather than the physical location of every device using the prefix. It is therefore useful evidence, but it should not be treated as a definitive GeoIP record.

RFC 9632 explicitly distinguishes operator-published geofeed data from geographic hints in RIR objects. Under its discovery rules, consumers must disregard other geographic hints for the relevant address range, such as country: or geoloc:, when a geofeed is specified. This does not guarantee that every commercial provider will accept the declared location without its own checks.

The distinction matters for multinational networks, cloud infrastructure, leased address space and anycast services. The registered organisation may be in one country while users or infrastructure are distributed across several others.

Which Data Source Is My Service Actually Using: RIR, Geofeed or a GeoIP Vendor?

Source	What it can show	Updated by the transfer?	How to correct it
RIR RDAP/WHOIS	Registered holder, contacts, status and registry country data	The holder record is updated through the RIR process	Use the relevant RIR or sponsoring LIR procedure
Geofeed	Operator-declared country, region and city by prefix	Not automatically	Publish an RFC 8805 geofeed and reference it through the registry’s supported mechanism
BGP	Observed announcements, prefixes and origin ASNs	No; routing must be reviewed separately	Coordinate any required announcement changes, IRR objects and RPKI ROAs
Commercial GeoIP	Vendor estimate of user or network location	Not necessarily; providers process evidence independently	Provide consistent evidence through each vendor’s correction channel
Online service	The service’s own interpretation, API result or licensed database snapshot	Not automatically	Identify its data supplier or use the service’s reporting process

Step 1: Verify the Transfer and Registry Record

Start with the authoritative registration data for the prefix. Confirm that the transfer is complete, the recipient organisation is displayed correctly and the abuse and administrative contacts are current. If the block is subdivided for customers, make sure any permitted more-specific assignments or suballocations are also accurate.

Do not change a registry country field solely to influence a consumer website. The record should accurately reflect the purpose defined by the relevant RIR. An inconsistent or misleading registry object can make correction requests less credible rather than more effective.

Step 2: How to Create a Geofeed File (RFC 8805 Example)

RFC 8805 defines a CSV format for publishing IP location data: ip_prefix,alpha2code,region,city,postal_code. Use ISO 3166-1 alpha-2 country codes and ISO 3166-2 region codes. Only the prefix is required; leave unknown location fields empty. The postal-code field is deprecated and should remain empty.

Choose prefixes that accurately represent deployment. If parts of a /20 serve different countries, publish more-specific entries for those parts. There is no need to split a block further when the same location accurately applies throughout it.

Host the file at a stable HTTPS URL. Use the relevant registry’s supported mechanism to reference it. For RPSL objects, RFC 9632 prefers geofeed: where supported and also defines the remarks: Geofeed form.

RFC 9877 defines how geofeed links can be exposed through RDAP. Follow the relevant RIR’s procedure; do not assume that every registry offers the same field or interface.

Example geofeed entries

192.0.2.0/24,DE,DE-BE,Berlin,  

198.51.100.0/25,ES,ES-M,Madrid,  

198.51.100.128/25,ES,ES-B,Barcelona,

These are documentation prefixes; replace them with authorised networks and verified locations. Save the actual CSV as UTF-8 with CRLF line endings and one record per line, without a header row. The trailing commas preserve the empty postal-code field. The /25 entries describe location granularity, not a requirement to announce /25 routes separately in BGP.

Step 3: Do BGP and RPKI Affect IP Geolocation?

A correct geofeed can still conflict with the network’s live operation. Verify that the observed announcements match the agreed deployment and that the authorised origin ASN is visible. This may be the recipient’s ASN or an ASN operated by its hosting, transit or other authorised network provider. A transfer does not, by itself, require a change of origin ASN.

Coordinate the withdrawal of obsolete or unauthorised announcements where necessary. Confirm that the appropriate RPKI Route Origin Authorisations (ROAs) cover the intended origin ASN and announced prefix lengths, maintain accurate IRR route objects, and review reverse DNS and abuse contacts.

BGP does not prove a user’s physical country, and a ROA authorises route origination rather than geographic location. An ASN can operate globally, and traffic can traverse remote upstreams or Internet exchanges. An unchanged origin ASN or multiple origin ASNs are not automatically errors; investigate whether the announcements are authorised and consistent with the deployment described in the correction request.

RPKI updates pass through repository publication, validator refresh and router-cache synchronisation. RFC 8210 specifies a recommended default of 3,600 seconds (one hour) for a router to poll its RPKI cache; a cache notification can prompt an earlier query. This is not an end-to-end propagation deadline. Check the ROA’s publication and the observed validation state rather than assuming that every network will reflect the change within one or two hours. If a route is invalid, check the authorised ASN and prefix length as well as possible stale data.

An IRR change also has several stages: acceptance by the authoritative registry, replication to any mirrors, and use by an upstream provider’s route-filtering system. Their schedules differ. Confirm the object in the authoritative IRR, then check which database and filter-refresh process the transit provider actually uses.

Step 4: Submit Corrections to GeoIP Providers

Once you have verified the public records and deployment, check several databases using addresses from each deployed subprefix. Record the returned country, region, city, organisation and connection type where available. Prioritise providers that are known to supply the affected product; correcting an unrelated lookup site may have no impact.

MaxMind provides an official GeoIP location correction form for an IP or network. It reviews requests against multiple data sources and does not guarantee acceptance. IPinfo also maintains an IP geolocation correction channel. Submit the exact CIDR, verified deployment location, company-domain contact and supporting geofeed or registry links, following each provider’s requirements.

If Google products show the wrong country, use Google’s IP problem reporting process. Google states that updates may take more than a month and that it does not provide an individual follow-up. A submitted request is therefore not an immediate correction.

Step 5: How to Check If the Geolocation Fix Worked

A public GeoIP demo does not necessarily use the same database version, product tier or custom rules as the service experiencing the problem. Test the actual search engine, payment gateway, advertising platform, content service or customer application affected by the error.

Use several addresses from the prefix and repeat tests from fresh sessions where appropriate. Confirm which public IP the service actually sees, especially when a VPN, proxy or NAT gateway is involved. Some services also consider account settings, device location and previous activity, so an unexpected country is not always caused by their GeoIP database.

Keep dated screenshots or API responses, the exact tested IPs and correction-ticket references. This creates a usable escalation record and helps distinguish an outdated vendor database from a cached application result or another location signal.

How Long Does a GeoIP Correction Take?

There is no universal update period. A provider must first receive or discover the new evidence, validate it and include any accepted correction in its data. Its customer must then download a new database release or obtain a refreshed API result. A downstream platform may apply additional rules or retain older data.

Routing and registry checks support an orderly transfer, but their completion does not set the timetable for a GeoIP correction.

Layer	When the change becomes visible	What can delay it
RPKI ROA	After publication, validator processing and router-cache synchronisation; no universal end-to-end deadline	Repository errors, failed validation, refresh intervals or stale caches
IRR route object	After registry acceptance; mirrors and upstream filters update separately	Authentication review, replication schedules or upstream provisioning
GeoIP provider correction	After review and publication; Google warns that its updates may take more than one month	Conflicting evidence, review requirements or release schedules
API-client cache	When the application refreshes its stored result; there is no standard cache lifetime	Custom TTL, additional cache layers or results stored without automatic refresh
Downloaded GeoIP database	When the customer downloads and deploys a build containing the correction	Update schedules, failed downloads or delayed deployment

Allow for provider-specific review and downstream update delays rather than assuming that the country will change when the RIR approves the transfer. Some services may take weeks or longer. Monitor progress by provider and avoid promising customers a single completion date.

A vendor API and a downloaded database behave differently. An API may expose corrected data once it is available in that product, but the customer application may retain its previous answer according to its cache policy. A local database remains unchanged until the customer downloads and deploys another build.

MaxMind’s database-update guidance recommends automation and checking periodically throughout the day for releases, including off-schedule updates. Its sample twice-weekly cron schedule is an example of customer-side download automation, not a correction deadline. If the vendor’s own lookup is correct but the end service is still wrong, check the database product, build date, cache policy and other location signals used by that service.

Not Sure Your Geofeed, Routing and Registry Data Agree?

Send us the prefix and we’ll check the public RIR record, geofeed, BGP origin, RPKI status and the main GeoIP databases for conflicting signals.

Check My IP Block

Common Reasons a Correction Fails

the transfer is complete, but an obsolete more-specific record still identifies the previous organisation;
the geofeed is unreachable, malformed or not discoverable through the registry or the provider’s submission process;
one geofeed row assigns a mixed-location block to a single country;
the observed announcements conflict with the deployment described in the request;
the requested country reflects the company’s registered address rather than the actual deployment;
only one GeoIP vendor was contacted while the affected service uses another source;
the provider has corrected its data, but the downstream service still uses an older database or cached result;
the service’s displayed country is influenced by account or device settings rather than IP geolocation alone.

GeoIP Checklist Before and After an IPv4 Transfer

Before the transfer

check representative addresses across the full prefix and relevant more-specifics in several GeoIP databases;
record current RIR, BGP, RPKI, IRR and geofeed data;
define the intended deployment locations and the prefixes to which they apply;
agree on technical cooperation, authorised origin ASNs and any required announcement withdrawals or changes;
review IP reputation as a separate risk; accurate geolocation does not establish that a block is free of abuse history or suitable for a particular service.

After the transfer

verify the recipient and contacts in RIR records;
publish or update an accurate geofeed and reference it using the registry’s supported mechanism;
verify BGP, RPKI, IRR and reverse-DNS records and complete any required changes;
submit GeoIP corrections with consistent evidence where location data is outdated;
test the affected customer-facing services and track unresolved discrepancies by provider.

Organisations planning to buy IPv4 address space should include geolocation verification in the transfer plan. Document the actual deployment, the services that need to recognise it correctly and the acceptable lead time before deployment.