- Created by Joakim von Brandis, last modified by Developers Edit User on Dec 08, 2022
Are you looking to use the public PassiveDNS database API? Please see the simplified integration guide for external users.
The Swagger API documentation is always up-to-date and lets you try out any query with your user session or an API-key.
Introduction
Passive Domain Name System (PDNS) is a database consisting of domain names as used by different IPs. The PDNS service exposes three different endpoints that can be used to search on these records. These are described below. Each record is marked with a TLP, indicating the sensitivity of the record. Records with TLP white are publicly accessible.
Finally, there is an endpoint that can be used to submit PDNS records in bulk. This is described towards the bottom of this document.
Concept | Description |
---|---|
DNS | |
Passive DNS | DNS records that the PDNS service collects |
TLP | |
Submission | Refers to a DNS record submitted to the PDNS REST API |
Simple query
To perform a simple PDNS query, use
https://api.mnemonic.no/pdns/v3/<query>?parameters
where "<query>" is a DNS query or answer string to lookup, for example
https://api.mnemonic.no/pdns/v3/cnn.com?rrType=A
Query parameters
There are multiple query parameters that can be passed along:
Parameter | Possible values | Default value | Description |
---|---|---|---|
limit | [0-maxint] | 25 | Limit the number of returned values. A value of "0" means that the client explicitly requests an unlimited result set. |
offset | [0-maxint] | 0 | Skip the initial <offset> number of values in the result set. Together with the parameter "limit", this allows a client to perform pagination of the results. |
rrClass | Any DNS record class | <any> | The most probable record class is in. By default, this field is not filtered. |
rrType | Any DNS record type | <any> | The most used record classes are a, aaaa, ptr, cname and mx, but any supported DNS record name can be used here. By default, this field is not filtered. |
The following options are only meaningful for use with private PDNS data | |||
customerID | [0-maxint] | <any> | If the user is authenticated, and has access to private PDNS data, this filter can limit which customers to return data for. Only records from customers the user has access to will be affected by this filter. Public data from other customers are "anonymous", and can be filtered out using the parameter "includeAnonymous". |
tlp | white,green,amber,red | <any> | If set, limit the returned records by TLP. For customers which the user does not have access to private data, only TLP white records will be returned. If the user is not authenticated, then also will only TLP white records be returned. |
aggregate | true,false | true | The PDNS service contains data from multiple customers, and PDNS records are stored per customer. By default, any data returned by the service will be aggregated into one record per query/answer tuple. If aggregate is set to false, the service will return each non-anonymous record separately. |
includeAnonymous | true,false | true | If true (default), any public data in the PDNS service will be included in the query results. If set to false, only data from customers the user has access to will be included. |
Result format
The result format is JSON, and consists of a result container, and a number of results. The results are sorted based on the lastSeenTImestamp in descending order.
The result container has the following format
{ "responseCode": 200, # the response code. Normal responses should return HTTP code 200 "count": 20, # the total number of matching records. If this value is lower than the imposed limit, the result set is truncated! "limit": 25, # the limit imposed on the query results (default is 25). Use the limit parameter in the request to set a higher limit. "offset": 0, # the offset applied on the query results (default i 0) "currentPage": 1, # the current "page" (calculated from limit/offset) "size": 20, # the size of the current result set (should be same as count or limit) "data": [], # the list of query result objects "messages": [], # any server messages "metaData": {}, # any server metadata }
Each query result has the following format:
{ "rrclass": "in", # the DNS record class "rrtype": "a", # the DNS record type "query": "cnn.com.", # the DNS record query part "answer": "157.166.255.19", # the DNS record answer part "firstSeenTimestamp": 1340308340000, # the first registered timestamp for this record "lastSeenTimestamp": 1377520248000, # the last registered timestamp for this record "maxTtl": 300, # the maximum TTL observed for this record "minTtl": 300, # the minimum TTL observed for this record "times": 675, # the number of times this record has been observed "tlp": "white", # the TLP of this record. Public records have TLP "white" "customer": null, # the customer owning this record. If null, this is an aggregated or anonymous record. "createdTimestamp": 0, # not in use "lastUpdatedTimestamp": 0, # not in use }
Examples
Query | Description |
---|---|
https://api.mnemonic.no/pdns/v3/cnn.com | Query for any record concerning the domain "cnn.com" |
https://api.mnemonic.no/pdns/v3/cnn.com?limit=0 | Query for any record concerning the domain "cnn.com". |
https://api.mnemonic.no/pdns/v3/cnn.com?rrType=a | Query for a-records for the domain "cnn.com" |
https://api.mnemonic.no/pdns/v3/cnn.com?rrType=A&rrType=mx | Query for a-records and mx-records for the domain "cnn.com" |
https://api.mnemonic.no/pdns/v3/157.166.255.19 | Query for any record for the IPv4 address "157.166.255.19" |
https://api.mnemonic.no/pdns/v3/2a04:4e42:400::323 | Query for any record for the IPv6 address "2a04:4e42:400::323" |
Output in COFF
The PDNS API also supports the PassiveDNS Common Output Format, as specified in https://datatracker.ietf.org/doc/draft-dulaunoy-dnsop-passive-dns-cof/
To retrieve results in the Common Output Format, please use the endpoint https://api.mnemonic.no/pdns/v3/cof/<query>
The query input format is the same as for the simple query endpoint, as described above.
JSON query format
An alternative to a simple GET-string is a POST query with a JSON query format to the URL
https://api.mnemonic.no/pdns/v3/search
Example
curl -X POST https://api.mnemonic.no/pdns/v3/search -d '{ "query": "cnn.com", "rrClass": [ "in" ], "rrType": [ "a" ], "limit": 0, "offset": 0 }
The POST-query is currently permitting the same parameters as the GET query, but new future parameters or advanced/nested parameters may be added only to the JSON query format as an "advanced format".
Querying if a PDNS record has been seen
If you only need to know whether a record matches the search criteria, you can use the seen endpoint
https://api.mnemonic.no/pdns/v3/<query>/seen?parameters
where "<query>" is a DNS query or answer string to lookup, for example.
curl https://api.mnemonic.no/pdns/v3/<query>/seen?ignoreOwnRecords=false
This endpoint will return a boolean, true or false, indicating whether a matching record exists.
Query parameters
There are multiple query parameters that can be passed along:
Parameter | Possible values | Default value | Description |
---|---|---|---|
tlp | white, green, amber, red | white | Only search in records matching any of the specified TLPs. Multiple values can be submitted. |
ignoreOwnRecords | true, false | true | Whether to ignore user's customer's own records, defaults to true |
Result format
The result format is JSON, and consists of a result container, and a number of results.
The result container has the following format
{ "responseCode": 200, # the response code. Normal responses should return HTTP code 200 "count": 0, # will always be 0 for this endpoint "limit": 0, # will always be 0 for this endpoint "offset": 0, # the offset applied on the query results (default i 0) "size": 20, # the size of the current result set (should be same as count or limit) "data": { # the query result "exists": false # a boolean indicating whether any matching record(s) was found }, "messages": [], # any server messages "metaData": {}, # any server metadata }
Authenticated queries
The PDNS API is publicly available, and does not require authentication.
However, unauthenticated queries are limited to see only public data (TLP white), and are limited to 1000 requests per day (currently).
To access private data (granted that mnemonic is collecting PDNS data from your customer), you need to use an authenticated query.
Also, users with an extended PDNS resource limit must use authenticated queries to be able to use their quota.
To authenticate your query, you need to request an API key from mnemonic.
To use that API-key, add a HTTP header to your request:
Argus-API-Key: 1234/1/abcd1234ef012
Example:
curl -H "Argus-API-Key: 1234/1/abcd1234ef012" -X GET https://api.mnemonic.no/pdns/v3/cnn.com
Resource limits
All users are subject to resource limitations.
Unauthenticated users are limited to 100 requests per minute, and 1000 requests per day.
Authenticated users are limited according to their agreement with mnemonic.
If you hit the resource limit, Argus will return a 402 error, with the following JSON response:
{ "responseCode": 402, "data": null, "messages": [{ "message": "Resource limit exceeded", "messageTemplate": "resource.limit.exceeded", "type": "ACTION_ERROR" }], "metaData": { "millisUntilResourcesAvailable": 558 } }
The resource limit is calculated both per minute and per day.
If you reach the per-minute (short-term) resource limit, you will typically be rejected for a short period, to save resources on our end.
If you reach the per-day (long-term) resource limit, you will typically be rejected for a period up to 24 hours. This means you have exhausted the data quota granted by mnemonic.
If you find yourself reaching the per-day resource limit a lot, you may want to request a higher resource quota from mnemonic
Please use the metadata key "millisUntilResourcesAvailable" to let your client back off gracefully.
Submit PassiveDNS records in bulk operation
PassiveDNS records can be submitted to the collector by using the POST endpoint /record in the API and will accept a list of multiple records. The minimum information in these records are the fields 'query', 'answer', 'firstSeen', 'times', 'minTtl', 'maxTtl', 'tlp' and 'recordType', or the request will be considered invalid. If not given the field recordClass defaults to 'IN', while lastSeen will be the submission time unless a timestamp is given.
If any of the records are invalid, the default behaviour is to fail the entire request and return the reasons why to the user. This can however be overwritten by using the field 'ignoreOnFailed', in which case the valid records are submitted while the invalid ones are ignored, though the response will contain a list of those that failed the validation and why.
It should also be noted that unless 'query' or 'answer' is written as IP addresses, they are expected to end with a period(.). If one is missing then the endpoint will add one before they are stored.
Below is a minimal request:
curl -X POST -H "Argus-API-Key: my/api/key" -H "Content-Type: application/json" https://api.mnemonic.no/pdns/v3/record -d '{ "records": [ { "query": "192.168.10.15", "answer": "192.168.10.16", "firstSeen": 1576589261000, "times": 12, "minTtl": 1, "maxTtl": 2, "tlp": "white", "recordType": "aaaa" } ] }'
You can check the swagger documentation for a more detailed description of the endpoint and the values that can be used in the Json object. |
---|
Another request using strings for 'query' and 'answer', and an ISO-8601 timestamp for 'firstSeen'
curl -X POST -H "Argus-API-Key: my/api/key" -H "Content-Type: application/json" https://api.mnemonic.no/pdns/v3/record -d '{ "records": [ { "query": "mnemonic.no", "answer": "not-mnemonic.no.", "firstSeen": "2016-11-30T15:47:00Z", "times": 12, "minTtl": 1, "maxTtl": 2, "tlp": "white", "recordType": "aaaa" } ] }'
You can check the swagger documentation for a more detailed description of the endpoint and the values that can be used in the Json object. |
---|
- No labels