Skip to content

massimocandela/geofeed-finder

Repository files navigation

geofeed-finder

This utility discovers and retrieves geofeed files from whois data. Additionally, it validates the ownership of the prefixes, manages the cache, and validates the ISO codes. See RFC9092.

To use the compiled version (linux, mac, windows), see releases. Otherwise, you can just download the code and do npm ci and npm run serve to run it.

How to set up a geofeed

Add a remark/comment in your inetnum/NetRange as follows:

Geofeed https://url_to_geofeed/file.csv

Example of result

$whois 209.212.224.1

NetRange:       209.212.224.0 - 209.212.239.255
CIDR:           209.212.224.0/20
NetName:        NTTA-209-212-224
NetHandle:      NET-209-212-224-0-1
Parent:         NET209 (NET-209-0-0-0-0)
NetType:        Direct Allocation
OriginAS:       AS2914
Organization:   NTT America, Inc. (NTTAM-1)
RegDate:        1998-04-17
Updated:        2020-12-18
Comment:        Geofeed https://geo.ip.gin.ntt.net/geofeeds/geofeeds.csv

Geofeed-finder usage examples

If you just added a geofeed link in your inetnum/NetRange and you want to test that everything is fine:

  • Run the binary ./geofeed-finder-linux-x64 -t YOUR_PREFIX

The -t option is not a prefix to geolocation lookup mechanism, but a test that your geofeed file is linked and constructed properly. This command will (1) find the parent inetnum for your prefix; (2) return all the geofeeds available for it; and (3) validate CSV format and ISO codes and report errors.

If you want to retrieve all the geofeeds in a RIR:

  • Run the binary ./geofeed-finder-linux-x64 -i ripe
  • See the final geofeed file in result.csv

You can select multiple RIRs: ./geofeed-finder-linux-x64 -i ripe,apnic

If you want to discover geofeeds across all RIRs:

  • Run the binary ./geofeed-finder-linux-x64
  • See the final geofeed file in result.csv

The final geofeed file is a file containing all the geofeeds discovered in whois data. Each entry is a prefix or IP which has been selected according to the draft (e.g. accepted only if contained by parent inetnum, priority to longest prefix match, etc.)

The application accepts the following parameters:

Parameter Description
-i Include RIRs (comma-separated list). Possible values are ripe, apnic, lacnic, afrinic, and arin.
-v Show version number.
-o Output file.
-t Test specific inetnum using RDAP.
-s Silent mode, don't print errors.
-c Whois cache validity (in days). Do not set, use the default instead.
-g Geofeed file cache validity (in days). Do not set, use the default instead.
-k Keep entries with invalid ISO codes. Not recommended.
-u Keep invalid subdivisions (accept invalid ISO regions/subdivisions, but keep validating the rest). Not recommended.
-r Remove invalid subdivisions but keep the rest of the geofeed if valid.
-z Include Zip codes. Not recommended. Zip codes are deprecated in geofeed and by default are excluded from the output.
-d Download timeout. Interrupt downloading a geofeed file after seconds. Default: 10 seconds.
-p Do not fetch arin sub allocations. You will save considerable time but have a potentially partial output.
-q Detect ARIN's sub allocations locally instead of downloading a dump file.
-a A comma-separated list of address families. Default: 4,6.
-l Cache directory. Default: .cache.

Use -h for more options. See here more information about -k, -u, and -r.

Downloading data from ARIN whois takes longer. This is because the other RIRs publicly provide anonymized bulk whois data. Instead, ARIN requires authorization to access bulk whois data. If you have such authorization, soon there will be an option to use ARIN bulk data, otherwise rdap is used (default, which doesn't require authorization.)

WARNING: do not remove the cache at each run and do not set cache vailidity values too small. If you do that, the risk that you get blocked by the data repositories is great.

Logs

All the logs, including ISO codes error are reported in logs/.

Use geofeed-finder in your code

Install it:

npm install geofeed-finder

Import it:

import GeofeedFinder from "geofeed-finder";

Use it:

const options = {
    include: ["ripe", "apnic"], // The RIRs to explore (default: ripe, apnic, lacnic, afrinic, arin),
    whoisCacheDays: 3, // Cache days for whois data (default: 3)
    geofeedCacheDays: 7, // Cache days for geofeed files without cache headers set (default: 7)
    af: [4, 6], // Address family (default, both 4 and 6)
    includeZip: true | false, // Allow for zip codes in the final output (default: false)
    silent: true | false, // Don't log in console (default: false)
    keepNonIso: true | false, // Don't validate ISO codes (default: false)
    keepInvalidSubdivisions: true | false, // Don't validate ISO codes of the subdivisions (default: false)
    removeInvalidSubdivisions: true | false, // Remove invalid subdivisions but keep the rest of the geofeed if valid (default: false)
    skipSuballocations: true | false, // Skip fetching ARIN sub allocations
    test: "ip/prefix", // Test specific ip/prefix using RDAP
    output: "result.csv", // Output file (default: "result.csv")
    downloadTimeout: 5 // Interrupt downloading a geofeed file after seconds (default: 10)
};

new GeofeedFinder(options) // The options dict is optional, you can just do new GeofeedFinder()
    .getGeofeeds()
    .then(geofeeds => { 
        // Do something with the geofeeds 
        // An array of objects { prefix, country, region, city }
    });