Bots

Contents

General remarks

By default all of the bots are started when you start the whole botnet, however there is a possibility to disable a bot. This means that the bot will not start every time you start the botnet, but you can start and stop the bot if you specify the bot explicitly. To disable a bot, add the following to your runtime.conf: “enabled”: false. Be aware that this is not a normal parameter (like the others described in this file). It is set outside of the parameters object in runtime.conf. Check out Configuration and Management for an example.

There are two different types of parameters: The initialization parameters are need to start the bot. The runtime parameters are needed by the bot itself during runtime.

The initialization parameters are in the first level, the runtime parameters live in the parameters sub-dictionary:

bot-id:
  parameters:
      runtime parameters...
  initialization parameters...

For example:

abusech-feodo-domains-collector:
  parameters:
    provider: Abuse.ch
    name: Abuse.ch Feodo Domains
    http_url: http://example.org/feodo-domains.txt
  name: Generic URL Fetcher
  group: Collector
  module: intelmq.bots.collectors.http.collector_http
  description: collect report messages from remote hosts using http protocol
  enabled: true
  run_mode: scheduled

This configuration resides in the file runtime.yaml in your IntelMQ’s configuration directory for each configured bot.

Initialization parameters

  • name and description: The name and description of the bot. See also intelmqctl list --configured bots.

  • group: Can be “Collector”, “Parser”, “Expert” or “Output”. Only used for visualization by other tools.

  • module: The executable (should be in $PATH) which will be started.

  • enabled: If the parameter is set to true (which is NOT the default value if it is missing as a protection) the bot will start when the botnet is started (intelmqctl start). If the parameter was set to false, the Bot will not be started by intelmqctl start, however you can run the bot independently using intelmqctl start <bot_id>. Check Configuration and Management for more details.

  • run_mode: There are two run modes, “continuous” (default run mode) or “scheduled”. In the first case, the bot will be running forever until stopped or exits because of errors (depending on configuration). In the latter case, the bot will stop after one successful run. This is especially useful when scheduling bots via cron or systemd. Default is continuous. Check Configuration and Management for more details.

Common parameters

Feed parameters

Common configuration options for all collectors.

  • name: Name for the feed (feed.name). In IntelMQ versions smaller than 2.2 the parameter name feed is also supported.

  • accuracy: Accuracy for the data of the feed (feed.accuracy).

  • code: Code for the feed (feed.code).

  • documentation: Link to documentation for the feed (feed.documentation).

  • provider: Name of the provider of the feed (feed.provider).

  • rate_limit: time interval (in seconds) between fetching data if applicable.

HTTP parameters

Common URL fetching parameters used in multiple bots.

  • http_timeout_sec: A tuple of floats or only one float describing the timeout of the HTTP connection. Can be a tuple of two floats (read and connect timeout) or just one float (applies for both timeouts). The default is 30 seconds in default.conf, if not given no timeout is used. See also https://requests.readthedocs.io/en/master/user/advanced/#timeouts

  • http_timeout_max_tries: An integer depicting how often a connection is retried, when a timeout occurred. Defaults to 3 in default.conf.

  • http_username: username for basic authentication.

  • http_password: password for basic authentication.

  • http_proxy: proxy to use for HTTP

  • https_proxy: proxy to use for HTTPS

  • http_user_agent: user agent to use for the request.

  • http_verify_cert: path to trusted CA bundle or directory, false to ignore verifying SSL certificates, or true (default) to verify SSL certificates

  • ssl_client_certificate: SSL client certificate to use.

  • ssl_ca_certificate: Optional string of path to trusted CA certificate. Only used by some bots.

  • http_header: HTTP request headers

Cache parameters

Common Redis cache parameters used in multiple bots (mainly lookup experts):

  • redis_cache_host: Hostname of the Redis database.

  • redis_cache_port: Port of the Redis database.

  • redis_cache_db: Database number.

  • redis_cache_ttl: TTL used for caching.

  • redis_cache_password: Optional password for the Redis database (default: none).

Collector Bots

Multihreading is disabled for all Collectors, as this would lead to duplicated data.

AMQP

Requires the pika python library, minimum version 1.0.0.

Information

  • name: intelmq.bots.collectors.amqp.collector_amqp

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: collect data from (remote) AMQP servers, for both IntelMQ as well as external data

Configuration Parameters

  • Feed parameters (see above)

  • connection_attempts: The number of connection attempts to defined server, defaults to 3

  • connection_heartbeat: Heartbeat to server, in seconds, defaults to 3600

  • connection_host: Name/IP for the AMQP server, defaults to 127.0.0.1

  • connection_port: Port for the AMQP server, defaults to 5672

  • connection_vhost: Virtual host to connect, on an HTTP(S) connection would be http:/IP/<your virtual host>

  • expect_intelmq_message: Boolean, if the data is from IntelMQ or not. Default: false. If true, then the data can be any Report or Event and will be passed to the next bot as is. Otherwise a new report is created with the raw data.

  • password: Password for authentication on your AMQP server

  • queue_name: The name of the queue to fetch data from

  • username: Username for authentication on your AMQP server

  • use_ssl: Use ssl for the connection, make sure to also set the correct port, usually 5671 (true/false)

Currently only fetching from a queue is supported can be extended in the future. Messages will be acknowledge at AMQP after it is sent to the pipeline.

API

Information

  • name: intelmq.bots.collectors.api.collector

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: collect report messages from an HTTP or Socket REST API

Configuration Parameters

  • Feed parameters (see above)

  • port: Optional, integer. Default: 5000. The local port, the API will be available at.

  • use_socket: Optional, boolean. Default: false. If true, the socket will be opened at the location given with socket_path.

  • socket_path: Optional, string. Default: /tmp/imq_api_default_socket

The API is available at /intelmq/push if the HTTP interface is used (default). The tornado library is required.

Generic URL Fetcher

Information

  • name: intelmq.bots.collectors.http.collector_http

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: collect report messages from remote hosts using HTTP protocol

Configuration Parameters

  • Feed parameters (see above)

  • HTTP parameters (see above)

  • extract_files: Optional, boolean or list of strings. If it is true, the retrieved (compressed) file or archived will be uncompressed/unpacked and the files are extracted. If the parameter is a list for strings, only the files matching the filenames are extracted. Extraction handles gzipped files and both compressed and uncompressed tar-archives as well as zip archives.

  • http_url: location of information resource (e.g. https://feodotracker.abuse.ch/blocklist/?download=domainblocklist)

  • http_url_formatting: (bool|JSON, default: false) If true, {time[format]} will be replaced by the current time in local timezone formatted by the given format. E.g. if the URL is http://localhost/{time[%Y]}, then the resulting URL is http://localhost/2019 for the year 2019. (Python’s Format Specification Mini-Language is used for this.). You may use a JSON specifying time-delta parameters to shift the current time accordingly. For example use {“days”: -1} for the yesterday’s date; the URL http://localhost/{time[%Y-%m-%d]} will get translated to “http://localhost/2018-12-31” for the 1st Jan of 2019.

  • verify_pgp_signatures: bool, defaults to false. If true, signature file is downloaded and report file is checked. On error (missing signature, mismatch, …), the error is logged and the report is not processed. Public key has to be imported in local keyring. This requires the python-gnupg library.

  • signature_url: Location of signature file for downloaded content. For path http://localhost/data/latest.json this may be for example http://localhost/data/latest.asc.

  • signature_url_formatting: (bool|JSON, default: false) The same as http_url_formatting, only for the signature file.

  • gpg_keyring: string or none (default). If specified, the string represents path to keyring file, otherwise the PGP keyring file for current intelmq user is used.

Zipped files are automatically extracted if detected.

For extracted files, every extracted file is sent in its own report. Every report has a field named extra.file_name with the file name in the archive the content was extracted from.

HTTP Response status code checks

If the HTTP response’ status code is not 2xx, this is treated as error.

In Debug logging level, the request’s and response’s headers and body are logged for further inspection.

Generic URL Stream Fetcher

Information

  • name: intelmq.bots.collectors.http.collector_http_stream

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: Opens a streaming connection to the URL and sends the received lines.

Configuration Parameters

  • Feed parameters (see above)

  • HTTP parameters (see above)

  • strip_lines: boolean, if single lines should be stripped (removing whitespace from the beginning and the end of the line)

If the stream is interrupted, the connection will be aborted using the timeout parameter. No error will be logged if the number of consecutive connection fails does not reach the parameter error_max_retries. Instead of errors, an INFO message is logged. This is a measurement against too frequent ERROR logging messages. The consecutive connection fails are reset if a data line has been successfully transferred. If the consecutive connection fails reaches the parameter error_max_retries, an exception will be thrown and rate_limit applies, if not null.

The parameter http_timeout_max_tries is of no use in this collector.

Generic Mail URL Fetcher

Information

  • name: intelmq.bots.collectors.mail.collector_mail_url

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: collect messages from mailboxes, extract URLs from that messages and download the report messages from the URLs.

Configuration Parameters

  • Feed parameters (see above)

  • HTTP parameters (see above)

  • mail_host: FQDN or IP of mail server

  • mail_user: user account of the email account

  • mail_password: password associated with the user account

  • mail_port: IMAP server port, optional (default: 143 without SSL, 993 for SSL)

  • mail_ssl: whether the mail account uses SSL (default: true)

  • folder: folder in which to look for mails (default: INBOX)

  • subject_regex: regular expression to look for a subject

  • url_regex: regular expression of the feed URL to search for in the mail body

  • sent_from: filter messages by sender

  • sent_to: filter messages by recipient

  • ssl_ca_certificate: Optional string of path to trusted CA certificate. Applies only to IMAP connections, not HTTP. If the provided certificate is not found, the IMAP connection will fail on handshake. By default, no certificate is used.

The resulting reports contains the following special fields:

  • feed.url: The URL the data was downloaded from

  • extra.email_date: The content of the email’s Date header

  • extra.email_subject: The subject of the email

  • extra.email_from: The email’s from address

  • extra.email_message_id: The email’s message ID

  • extra.file_name: The file name of the downloaded file (extracted from the HTTP Response Headers if possible).

Chunking

For line-based inputs the bot can split up large reports into smaller chunks.

This is particularly important for setups that use Redis as a message queue which has a per-message size limitation of 512 MB.

To configure chunking, set chunk_size to a value in bytes. chunk_replicate_header determines whether the header line should be repeated for each chunk that is passed on to a parser bot.

Specifically, to configure a large file input to work around Redis’ size limitation set chunk_size to something like 384000000, i.e., ~384 MB.

Generic Mail Attachment Fetcher

Information

  • name: intelmq.bots.collectors.mail.collector_mail_attach

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: collect messages from mailboxes, download the report messages from the attachments.

Configuration Parameters

  • Feed parameters (see above)

  • extract_files: Optional, boolean or list of strings. See documentation of the Generic URL Fetcher for more details.

  • mail_host: FQDN or IP of mail server

  • mail_user: user account of the email account

  • mail_password: password associated with the user account

  • mail_port: IMAP server port, optional (default: 143 without SSL, 993 for SSL)

  • mail_ssl: whether the mail account uses SSL (default: true)

  • folder: folder in which to look for mails (default: INBOX)

  • subject_regex: regular expression to look for a subject

  • attach_regex: regular expression of the name of the attachment

  • attach_unzip: whether to unzip the attachment. Only extracts the first file. Deprecated, use extract_files instead.

  • sent_from: filter messages by sender

  • sent_to: filter messages by recipient

  • ssl_ca_certificate: Optional string of path to trusted CA certificate. Applies only to IMAP connections, not HTTP. If the provided certificate is not found, the IMAP connection will fail on handshake. By default, no certificate is used.

The resulting reports contains the following special fields:

  • extra.email_date: The content of the email’s Date header

  • extra.email_subject: The subject of the email

  • extra.email_from: The email’s from address

  • extra.email_message_id: The email’s message ID

  • extra.file_name: The file name of the attachment or the file name in the attached archive if attachment is to uncompress.

Generic Mail Body Fetcher

Information

  • name: intelmq.bots.collectors.mail.collector_mail_body

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: collect messages from mailboxes, forwards the bodies as reports. Each non-empty body with the matching content type is sent as individual report.

Configuration Parameters

  • Feed parameters (see above)

  • mail_host: FQDN or IP of mail server

  • mail_user: user account of the email account

  • mail_password: password associated with the user account

  • mail_port: IMAP server port, optional (default: 143 without SSL, 993 for SSL)

  • mail_ssl: whether the mail account uses SSL (default: true)

  • folder: folder in which to look for mails (default: INBOX)

  • subject_regex: regular expression to look for a subject

  • sent_from: filter messages by sender

  • sent_to: filter messages by recipient

  • ssl_ca_certificate: Optional string of path to trusted CA certificate. Applies only to IMAP connections, not HTTP. If the provided certificate is not found, the IMAP connection will fail on handshake. By default, no certificate is used.

  • content_types: Which bodies to use based on the content_type. Default: true/[‘html’, ‘plain’] for all: - string with comma separated values, e.g. [‘html’, ‘plain’] - true, false, null: Same as default value - string, e.g. ‘plain’

The resulting reports contains the following special fields:

  • extra.email_date: The content of the email’s Date header

  • extra.email_subject: The subject of the email

  • extra.email_from: The email’s from address

  • extra.email_message_id: The email’s message ID

Github API

Information

  • name: intelmq.bots.collectors.github_api.collector_github_contents_api

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: Collects files matched by regular expression from GitHub repository via the GitHub API. Optionally with GitHub credentials, which are used as the Basic HTTP authentication.

Configuration Parameters

  • Feed parameters (see above)

  • basic_auth_username: GitHub account username (optional)

  • basic_auth_password: GitHub account password (optional)

  • repository: GitHub target repository (<USER>/<REPOSITORY>)

  • regex: Valid regular expression of target files within the repository (defaults to .*.json)

  • extra_fields: Comma-separated list of extra fields from GitHub contents API.

Workflow

The optional authentication parameters provide a high limit of the GitHub API requests. With the git hub user authentication, the requests are rate limited to 5000 per hour, otherwise to 60 requests per hour.

The collector recursively searches for regex-defined files in the provided repository. Additionally it adds extra file metadata defined by the extra_fields.

The bot always sets the url, from which downloaded the file, as feed.url.

Fileinput

Information

  • name: intelmq.bots.collectors.file.collector_file

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: This bot is capable of reading files from the local file-system. This is handy for testing purposes, or when you need to react to spontaneous events. In combination with the Generic CSV Parser this should work great.

Configuration Parameters

  • Feed parameters (see above)

  • path: path to file

  • postfix: The postfix (file ending) of the files to look for. For example .csv.

  • delete_file: whether to delete the file after reading (default: false)

The resulting reports contains the following special fields:

  • feed.url: The URI using the file:// scheme and localhost, with the full path to the processed file.

  • extra.file_name: The file name (without path) of the processed file.

Chunking

Additionally, for line-based inputs the bot can split up large reports into smaller chunks.

This is particularly important for setups that use Redis as a message queue which has a per-message size limitation of 512 MB.

To configure chunking, set chunk_size to a value in bytes. chunk_replicate_header determines whether the header line should be repeated for each chunk that is passed on to a parser bot.

Specifically, to configure a large file input to work around Redis’ size limitation set chunk_size to something like 384000, i.e., ~384 MB.

Workflow

The bot loops over all files in path and tests if their file name matches postfix, e.g. `.csv`. If yes, the file will be read and inserted into the queue.

If delete_file is set, the file will be deleted after processing. If deletion is not possible, the bot will stop.

To prevent data loss, the bot also stops when no postfix is set and delete_file was set. This cannot be overridden.

The bot always sets the file name as feed.url

Fireeye

Information

  • name: intelmq.bots.collectors.fireeye.collector_fireeye

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: This bot is capable of collecting hashes and URLs from a Fireeye MAS appliance.

The Python library xmltodict is required to run this bot.

Configuration Parameters

  • Feed parameters (see above)

  • dns_name: DNS name of the target appliance.

  • request_duration: Length of the query in past eg. collect alerts from last 24hours/48hours.

  • http_username: Password for authentication.

  • http_password: Username for authentication.

Workflow

The bot collects all alerts which occurred during specified duration. After this we make a second call and check if there is additional information like domains and hashes available. After collecting the openioc data we send this information to the Fireeye parser.

Kafka

Requires the kafka python library.

Information

  • name: intelmq.bots.collectors.kafka.collector

Configuration parameters

  • topic: the kafka topic the collector should get messages from

  • bootstrap_servers: the kafka server(s) the collector should connect to. Defaults to localhost:9092

  • ssl_check_hostname: false to ignore verifying SSL certificates, or true (default) to verify SSL certificates

  • ssl_client_certificate: SSL client certificate to use.

  • ssl_ca_certificate: Optional string of path to trusted CA certificate. Only used by some bots.

Rsync

Requires the rsync executable

Information

  • name: intelmq.bots.collectors.rsync.collector_rsync

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: Bot download file by rsync and then load data from downloaded file. Downloaded file is located in var/lib/bots/rsync_collector.

Configuration Parameters

  • Feed parameters (see above)

  • file: Name of downloaded file.

  • rsync_path: Path to file. It can be “/home/username/directory” or “username@remote_host:/home/username/directory”

  • temp_directory: Path of a temporary state directory to use for rsync’d files. Optional. Default: /opt/intelmq/var/run/rsync_collector/.

MISP Generic

Information

  • name: intelmq.bots.collectors.misp.collector

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: collect messages from MISP, a malware information sharing platform server.

Configuration Parameters

  • Feed parameters (see above)

  • misp_url: URL of MISP server (with trailing ‘/’)

  • misp_key: MISP Authkey

  • misp_tag_to_process: MISP tag for events to be processed

  • misp_tag_processed: MISP tag for processed events, optional

Generic parameters used in this bot:

  • http_verify_cert: Verify the TLS certificate of the server, boolean (default: true)

Workflow This collector will search for events on a MISP server that have a to_process tag attached to them (see the misp_tag_to_process parameter) and collect them for processing by IntelMQ. Once the MISP event has been processed the to_process tag is removed from the MISP event and a processed tag is then attached (see the misp_tag_processed parameter).

NB. The MISP tags must be configured to be ‘exportable’ otherwise they will not be retrieved by the collector.

Request Tracker

Information

  • name: intelmq.bots.collectors.rt.collector_rt

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: Request Tracker Collector fetches attachments from an RTIR instance.

You need the rt-library >= 1.9 from nic.cz, available via pypi: pip3 install rt

This rt bot will connect to RT and inspect the given search_queue for tickets matching all criteria in search_*, Any matches will be inspected. For each match, all (RT-) attachments of the matching RT tickets are iterated over and within this loop, the first matching filename in the attachment is processed. If none of the filename matches apply, the contents of the first (RT-) “history” item is matched against the regular expression for the URL (url_regex).

Configuration Parameters

  • Feed parameters (see above)

  • HTTP parameters (see above)

  • extract_attachment: Optional, boolean or list of strings. See documentation of the Generic URL Fetcher parameter extract_files for more details.

  • extract_download: Optional, boolean or list of strings. See documentation of the Generic URL Fetcher parameter extract_files for more details.

  • uri: URL of the REST interface of the RT

  • user: RT username

  • password: RT password

  • search_not_older_than: Absolute time (use ISO format) or relative time, e.g. 3 days.

  • search_owner: owner of the ticket to search for (default: nobody)

  • search_queue: queue of the ticket to search for (default: Incident Reports)

  • search_requestor: the e-mail address of the requestor

  • search_status: status of the ticket to search for (default: new)

  • search_subject_like: part of the subject of the ticket to search for (default: Report)

  • set_status: status to set the ticket to after processing (default: open). false or null to not set a different status.

  • take_ticket: whether to take the ticket (default: true)

  • url_regex: regular expression of an URL to search for in the ticket

  • attachment_regex: regular expression of an attachment in the ticket

  • unzip_attachment: whether to unzip a found attachment. Only the first file in the archive is used. Deprecated in favor of extract_attachment.

The parameter http_timeout_max_tries is of no use in this collector.

The resulting reports contains the following special fields:

  • rtir_id: The ticket ID

  • extra.email_subject and extra.ticket_subject: The subject of the ticket

  • extra.email_from and extra.ticket_requestors: Comma separated list of the ticket’s requestor’s email addresses.

  • extra.ticket_owner: The ticket’s owner name

  • extra.ticket_status: The ticket’s status

  • extra.ticket_queue: The ticket’s queue

  • extra.file_name: The name of the extracted file, the name of the downloaded file or the attachments’ filename without .gz postfix.

  • time.observation: The creation time of the ticket or attachment.

Search

The parameters prefixed with search_ allow configuring the ticket search.

Empty strings and null as value for search parameters are ignored.

File downloads

Attachments can be optionally unzipped, remote files are downloaded with the http_* settings applied.

If url_regex or attachment_regex are empty strings, false or null, they are ignored.

Ticket processing

Optionally, the RT bot can “take” RT tickets (i.e. the user is assigned this ticket now) and/or the status can be changed (leave set_status empty in case you don’t want to change the status). Please note however that you MUST do one of the following: either “take” the ticket or set the status (set_status). Otherwise, the search will find the ticket every time and we will have generated an endless loop.

In case a resource needs to be fetched and this resource is permanently not available (status code is 4xx), the ticket status will be set according to the configuration to avoid processing the ticket over and over. For temporary failures the status is not modified, instead the ticket will be skipped in this run.

Time search

To find only tickets newer than a given absolute or relative time, you can use the search_not_older_than parameter. Absolute time specification can be anything parseable by dateutil, best use a ISO format.

Relative must be in this format: [number] [timespan]s, e.g. 3 days. timespan can be hour, day, week, month, year. Trailing ‘s’ is supported for all timespans. Relative times are subtracted from the current time directly before the search is performed.

Rsync

Information

  • name: intelmq.bots.collectors.rsync.collector_rsync

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: Syncs a file via rsync and reads the file.

Configuration Parameters

  • Feed parameters (see above)

  • file: The filename to process, combine with rsync_path.

  • temp_directory: The temporary directory for rsync, by default $VAR_STATE_PATH/rsync_collector. $VAR_STATE_PATH is /var/run/intelmq/ or /opt/intelmq/var/run/.

  • rsync_path: The path of the file to process

Shadowserver Reports API

The Cache is required to memorize which files have already been processed (TTL needs to be high enough to cover the oldest files available!).

Information

  • name: intelmq.bots.collectors.shadowserver.collector_reports_api

  • description: Connects to the Shadowserver API, requests a list of all the reports for a specific country and processes the ones that are new.

Configuration Parameters

  • country: The country you want to download the reports for

  • apikey: Your Shadowserver API key

  • secret: Your Shadowserver API secret

  • types: A list of strings or a string of comma-separated values with the names of report types you want to process. If you leave this empty, all the available reports will be downloaded and processed (i.e. ‘scan’, ‘drones’, ‘intel’, ‘sandbox_connection’, ‘sinkhole_combined’). The possible report types are equivalent to the file names given in the section Supported Reports of the Shadowserver parser.

  • Cache parameters (see in section Common parameters, the default TTL is set to 10 days)

The resulting reports contain the following special field:

  • extra.file_name: The name of the downloaded file, with fixed filename extension. The API returns file names with the extension .csv, although the files are JSON, not CSV. Therefore, for clarity and better error detection in the parser, the file name in extra.file_name uses .json as extension.

Shodan Stream

Requires the shodan library to be installed:

Information

  • name: intelmq.bots.collectors.shodan.collector_stream

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: Queries the Shodan Streaming API

Configuration Parameters

  • Feed parameters (see above)

  • HTTP parameters (see above). Only the proxy is used (requires shodan-python > 1.8.1). Certificate is always verified.

  • countries: A list of countries to query for. If it is a string, it will be spit by ,.

If the stream is interrupted, the connection will be aborted using the timeout parameter. No error will be logged if the number of consecutive connection fails does not reach the parameter error_max_retries. Instead of errors, an INFO message is logged. This is a measurement against too frequent ERROR logging messages. The consecutive connection fails are reset if a data line has been successfully transferred. If the consecutive connection fails reaches the parameter error_max_retries, an exception will be thrown and rate_limit applies, if not null.

TCP

Information

  • name: intelmq.bots.collectors.tcp.collector

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: TCP is the bot responsible to receive events on a TCP port (ex: from TCP Output of another IntelMQ instance). Might not be working on Python3.4.6.

Configuration Parameters

  • ip: IP of destination server

  • port: port of destination server

Response

TCP collector just sends an “Ok” message after every received message, this should not pose a problem for an arbitrary input. If you intend to link two IntelMQ instance via TCP, have a look at the TCP output bot documentation.

Alien Vault OTX

Information

  • name: intelmq.bots.collectors.alienvault_otx.collector

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: collect report messages from Alien Vault OTX API

Requirements

Install the library from GitHub, as there is no package in PyPi:

pip3 install -r intelmq/bots/collectors/alienvault_otx/REQUIREMENTS.txt

Configuration Parameters

  • Feed parameters (see above)

  • api_key: API Key

  • modified_pulses_only: get only modified pulses instead of all, set to it to true or false, default false

  • interval: if “modified_pulses_only” is set, define the time in hours (integer value) to get modified pulse since then, default 24 hours

Blueliv Crimeserver

Information

  • name: intelmq.bots.collectors.blueliv.collector_crimeserver

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: collect report messages from Blueliv API

For more information visit https://github.com/Blueliv/api-python-sdk

Requirements

Install the required library:

pip3 install -r intelmq/bots/collectors/blueliv/REQUIREMENTS.txt

Configuration Parameters

Calidog Certstream

A Bot to collect data from the Certificate Transparency Log (CTL) This bot works based on certstream library (https://github.com/CaliDog/certstream-python)

Information

  • name: intelmq.bots.collectors.calidog.collector_certstream

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: collect data from Certificate Transparency Log

Configuration Parameters

  • Feed parameters (see above)

ESET ETI

Information

  • name: intelmq.bots.collectors.eset.collector

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: collect data from ESET ETI TAXII server

For more information visit https://www.eset.com/int/business/services/threat-intelligence/

Requirements

Install the required cabby library:

pip3 install -r intelmq/bots/collectors/eset/REQUIREMENTS.txt

Configuration Parameters

  • Feed parameters (see above)

  • username: Your username

  • password: Your password

  • endpoint: eti.eset.com

  • time_delta: The time span to look back, in seconds. Default 3600.

  • collection: The collection to fetch.

McAfee openDXL

Information

  • name: intelmq.bots.collectors.opendxl.collector

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: collect messages via openDXL

Configuration Parameters

  • Feed parameters (see above)

  • dxl_config_file: location of the configuration file containing required information to connect $

  • dxl_topic: the name of the DXL topic to subscribe

Microsoft Azure

Iterates over all blobs in all containers in an Azure storage. The Cache is required to memorize which files have already been processed (TTL needs to be high enough to cover the oldest files available!).

This bot significantly changed in a backwards-incompatible way in IntelMQ Version 2.2.0 to support current versions of the Microsoft Azure Python libraries. azure-storage-blob>=12.0.0 is required.

Information

  • name: intelmq.bots.collectors.microsoft.collector_azure

  • lookup: yes

  • public: no

  • cache (redis db): 5

  • description: collect blobs from Microsoft Azure using their library

Configuration Parameters

  • Cache parameters (see above)

  • Feed parameters (see above)

  • connection_string: connection string as given by Microsoft

  • container_name: name of the container to connect to

Microsoft Interflow

Iterates over all files available by this API. Make sure to limit the files to be downloaded with the parameters, otherwise you will get a lot of data! The cache is used to remember which files have already been downloaded. Make sure the TTL is high enough, higher than not_older_than.

Information

  • name: intelmq.bots.collectors.microsoft.collector_interflow

  • lookup: yes

  • public: no

  • cache (redis db): 5

  • description: collect files from Microsoft Interflow using their API

Configuration Parameters

  • Feed parameters (see above)

  • api_key: API generate in their portal

  • file_match: an optional regular expression to match file names

  • not_older_than: an optional relative (minutes) or absolute time (UTC is assumed) expression to determine the oldest time of a file to be downloaded

  • redis_cache_* and especially redis_cache_ttl: Settings for the cache where file names of downloaded files are saved. The cache’s TTL must always be bigger than not_older_than.

Additional functionalities

  • Files are automatically ungzipped if the filename ends with .gz.

Stomp

Information

  • name: intelmq.bots.collectors.stomp.collector

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: collect messages from a stomp server

Requirements

Install the stomp.py library from PyPI:

pip3 install -r intelmq/bots/collectors/stomp/REQUIREMENTS.txt

Configuration Parameters

  • Feed parameters (see above)

  • exchange: exchange point

  • port: 61614

  • server: hostname e.g. “n6stream.cert.pl”

  • ssl_ca_certificate: path to CA file

  • ssl_client_certificate: path to client cert file

  • ssl_client_certificate_key: path to client cert key file

Twitter

Collects tweets from target_timelines. Up to tweet_count tweets from each user and up to timelimit back in time. The tweet text is sent separately and if allowed, links to pastebin are followed and the text sent in a separate report

Information

  • name: intelmq.bots.collectors.twitter.collector_twitter

  • lookup: yes

  • public: yes

  • cache (redis db): none

  • description: Collects tweets

Configuration Parameters

  • Feed parameters (see above)

  • target_timelines: screen_names of twitter accounts to be followed

  • tweet_count: number of tweets to be taken from each account

  • timelimit: maximum age of the tweets collected in seconds

  • follow_urls: list of screen_names for which URLs will be followed

  • exclude_replies: exclude replies of the followed screen_names

  • include_rts: whether to include retweets by given screen_name

  • consumer_key: Twitter API login data

  • consumer_secret: Twitter API login data

  • access_token_key: Twitter API login data

  • access_token_secret: Twitter API login data

API collector bot

Information

  • name: intelmq.bots.collectors.api.collector_api

  • lookup: no

  • public: no

  • cache (redis db): none

  • description: Bot for collecting data using API, you need to post JSON to /intelmq/push endpoint

example usage:

curl -X POST http://localhost:5000/intelmq/push -H 'Content-Type: application/json' --data '{"source.ip": "127.0.0.101", "classification.type": "system-compromise"}'

Configuration Parameters

  • Feed parameters (see above)

  • port: 5000

Parser Bots

Not complete

This list is not complete. Look at intelmqctl list bots or the list of parsers shown in the manager. But most parsers do not need configuration parameters.

TODO

AnubisNetworks Cyberfeed Stream

Information

  • name: intelmq.bots.parsers.anubisnetworks.parser

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: parsers data from AnubisNetworks Cyberfeed Stream

Description

The feed format changes over time. The parser supports at least data from 2016 and 2020.

Events with the Malware “TestSinkholingLoss” are ignored, as they are for the feed provider’s internal purpose only and should not be processed at all.

Configuration parameters

  • use_malware_familiy_as_classification_identifier: default: true. Use the malw.family field as classification.type. If false, check if the same as malw.variant. If it is the same, it is ignored. Otherwise saved as extra.malware.family.

Generic CSV Parser

Information

  • name: intelmq.bots.parsers.generic.parser_csv

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: Parses CSV data

Lines starting with ‘#’ will be ignored. Headers won’t be interpreted.

Configuration parameters

  • “columns”: A list of strings or a string of comma-separated values with field names. The names must match the IntelMQ Data Format field names. Empty column specifications and columns named “__IGNORE__” are ignored. E.g.

    "columns": [
         "",
         "source.fqdn",
         "extra.http_host_header",
         "__IGNORE__"
    ],
    

    is equivalent to:

    "columns": ",source.fqdn,extra.http_host_header,"
    

    The first and the last column are not used in this example.

    It is possible to specify multiple columns using the | character. E.g.

    "columns": "source.url|source.fqdn|source.ip"
    

    First, bot will try to parse the value as URL, if it fails, it will try to parse it as FQDN, if that fails, it will try to parse it as IP, if that fails, an error will be raised. Some use cases -

    • mixed data set, e.g. URL/FQDN/IP/NETMASK “columns”: “source.url|source.fqdn|source.ip|source.network”

    • parse a value and ignore if it fails “columns”: “source.url|__IGNORE__”

  • “column_regex_search”: Optional. A dictionary mapping field names (as given per the columns parameter) to regular expression. The field is evaluated using re.search. Eg. to get the ASN out of AS1234 use: {“source.asn”: “[0-9]*”}. Make sure to properly escape any backslashes in your regular expression (See also #1579).

  • “compose_fields”: Optional, dictionary. Create fields from columns, e.g. with data like this:

    # Host,Path
    example.com,/foo/
    example.net,/bar/
    

    using this compose_fields parameter:

    {"source.url": "http://{0}{1}"}
    

    You get:

    http://example.com/foo/
    http://example.net/bar/
    

    in the respective source.url fields. The value in the dictionary mapping is formatted whereas the columns are available with their index.

  • “default_url_protocol”: For URLs you can give a default protocol which will be pretended to the data.

  • “delimiter”: separation character of the CSV, e.g. “,”

  • “skip_header”: Boolean, skip the first line of the file, optional. Lines starting with # will be skipped additionally, make sure you do not skip more lines than needed!

  • time_format: Optional. If “timestamp”, “windows_nt” or “epoch_millis” the time will be converted first. With the default null fuzzy time parsing will be used.

  • “type”: set the classification.type statically, optional

  • “data_type”: sets the data of specific type, currently only “json” is supported value. An example

    {
        "columns": [ "source.ip", "source.url", "extra.tags"],
        "data_type": "{\"extra.tags\":\"json\"}"
    }
    

    It will ensure extra.tags is treated as json.

  • “filter_text”: only process the lines containing or not containing specified text, to be used in conjunction with filter_type

  • “filter_type”: value can be whitelist or blacklist. If whitelist, only lines containing the text in filter_text will be processed, if blacklist, only lines NOT containing the text will be processed.

    To process ipset format files use

    {
         "filter_text": "ipset add ",
         "filter_type": "whitelist",
         "columns": [ "__IGNORE__", "__IGNORE__", "__IGNORE__", "source.ip"]
    }
    
  • “type_translation”: If the source does have a field with information for classification.type, but it does not correspond to IntelMQ’s types, you can map them to the correct ones. The type_translation field can hold a dictionary, or a string with a JSON dictionary which maps the feed’s values to IntelMQ’s. Example:

    {"malware_download": "malware-distribution"}
    
  • “columns_required”: A list of true/false for each column. By default, it is true for every column.

Calidog Certstream

Information

  • name: intelmq.bots.parsers.calidog.parser_certstream

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: parsers data from Certificate Transparency Log

Description

For each domain in the leaf_cert.all_domains object one event with the domain in source.fqdn (and source.ip as fallback) is produced. The seen-date is saved in time.source and the classification type is other.

  • Feed parameters (see above)

ESET

Information

  • name: intelmq.bots.parsers.eset.parser

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: Parses data from ESET ETI TAXII server

Description

Supported collections:

  • “ei.urls (json)”

  • “ei.domains v2 (json)”

Cymru CAP Program

Information

  • name: intelmq.bots.parsers.cymru.parser_cap_program

  • public: no

  • cache (redis db): none

  • description: Parses data from Cymru’s CAP program feed.

Description

There are two different feeds available:

  • infected_$date.txt (“old”)

  • $certname_$date.txt (“new”)

The new will replace the old at some point in time, currently you need to fetch both. The parser handles both formats.

Old feed

As little information on the format is available, the mappings might not be correct in all cases. Some reports are not implemented at all as there is no data available to check if the parsing is correct at all. If you do get errors like Report … not implement or similar please open an issue and report the (anonymized) example data. Thanks.

The information about the event could be better in many cases but as Cymru does not want to be associated with the report, we can’t add comments to the events in the parser, because then the source would be easily identifiable for the recipient.

Cymru Full Bogons

http://www.team-cymru.com/bogon-reference.html

Information

  • name: intelmq.bots.parsers.cymru.parser_full_bogons

  • public: no

  • cache (redis db): none

  • description: Parses data from full bogons feed.

Github Feed

Information

  • name: intelmq.bots.parsers.github_feed.parser

  • description: Parses Feeds available publicly on GitHub (should receive from github_api collector)

Have I Been Pwned Callback Parser

Information

  • name: intelmq.bots.parsers.hibp.parser_callback

  • public: no

  • cache (redis db): none

  • description: Parses data from Have I Been Pwned feed.

Description

Parsers the data from a Callback of a Have I Been Pwned Enterprise Subscription.

Parses breaches and pastes and creates one event per e-mail address. The e-mail address is stored in source.account. classification.type is leak and classification.identifier is breach or paste.

HTML Table Parser

  • name: intelmq.bots.parsers.html_table.parser

  • public: yes

  • cache (redis db): none

  • description: Parses tables in HTML documents

Configuration parameters

  • “columns”: A list of strings or a string of comma-separated values with field names. The names must match the IntelMQ Data Format field names. Empty column specifications and columns named “__IGNORE__” are ignored. E.g.

    "columns": [
         "",
         "source.fqdn",
         "extra.http_host_header",
         "__IGNORE__"
    ],
    

    is equivalent to:

    "columns": ",source.fqdn,extra.http_host_header,"
    

    The first and the last column are not used in this example. It is possible to specify multiple columns using the | character. E.g.

    "columns": "source.url|source.fqdn|source.ip"
    

    First, bot will try to parse the value as URL, if it fails, it will try to parse it as FQDN, if that fails, it will try to parse it as IP, if that fails, an error will be raised. Some use cases -

    • mixed data set, e.g. URL/FQDN/IP/NETMASK “columns”: “source.url|source.fqdn|source.ip|source.network”

    • parse a value and ignore if it fails “columns”: “source.url|__IGNORE__”

  • “ignore_values”: A list of strings or a string of comma-separated values which will not considered while assigning to the corresponding fields given in columns. E.g.

    "ignore_values": [
         "",
         "unknown",
         "Not listed",
     ],
    

    is equivalent to:

    "ignore_values": ",unknown,Not listed,"
    

    The following configuration will lead to assigning all values to malware.name and extra.SBL except unknown and Not listed respectively.

    "columns": [
         "source.url",
         "malware.name",
         "extra.SBL",
    ],
    "ignore_values": [
         "",
         "unknown",
         "Not listed",
    ],
    

    Parameters columns and ignore_values must have same length

  • “attribute_name”: Filtering table with table attributes, to be used in conjunction with attribute_value, optional. E.g. class, id, style.

  • “attribute_value”: String. To filter all tables with attribute class=’details’ use

    "attribute_name": "class",
    "attribute_value": "details"
    
  • “table_index”: Index of the table if multiple tables present. If attribute_name and attribute_value given, index according to tables remaining after filtering with table attribute. Default: 0.

  • “split_column”: Padded column to be split to get values, to be used in conjunction with split_separator and split_index, optional.

  • “split_separator”: Delimiter string for padded column.

  • “split_index”: Index of unpadded string in returned list from splitting split_column with split_separator as delimiter string. Default: 0.

    E.g.

    "split_column": "source.fqdn",
    "split_separator": " ",
    "split_index": 1,
    

    With above configuration, column corresponding to source.fqdn with value [D] lingvaworld.ru will be assigned as “source.fqdn”: “lingvaworld.ru”.

  • “skip_table_head”: Boolean, skip the first row of the table, optional. Default: true.

  • “default_url_protocol”: For URLs you can give a default protocol which will be pretended to the data. Default: “http://”.

  • “time_format”: Optional. If “timestamp”, “windows_nt” or “epoch_millis” the time will be converted first. With the default null fuzzy time parsing will be used.

  • “type”: set the classification.type statically, optional

  • “html_parser”: The HTML parser to use, by default “html.parser”, can also be e.g. “lxml”, have a look at https://www.crummy.com/software/BeautifulSoup/bs4/doc/

Key-Value Parser

Information

  • name: intelmq.bots.parsers.key_value.parser

  • lookup: no

  • public: no

  • cache (redis db): none

  • description: Parses text lines in key=value format, for example FortiGate firewall logs.

Configuration Parameters

  • pair_separator: String separating key=value pairs, default ” “ (space).

  • kv_separator: String separating key and value, default =.

  • keys: Array of string->string, names of keys to propagate mapped to IntelMQ event fields. Example:

    "keys": {
        "srcip": "source.ip",
        "dstip": "destination.ip"
    }
    

    The value mapped to time.source is parsed. If the value is numeric, it is interpreted. Otherwise, or if it fails, it is parsed fuzzy with dateutil. If the value cannot be parsed, a warning is logged per line.

  • strip_quotes: Boolean, remove opening and closing quotes from values, default true.

Parsing limitations

The input must not have (quoted) occurrences of the separator in the values. For example, this is not parsable (with space as separator):

key="long value" key2="other value"

In firewall logs like FortiGate, this does not occur. These logs usually look like:

srcip=192.0.2.1 srcmac="00:00:5e:00:17:17"

McAfee Advanced Threat Defense File

Information

  • name: intelmq.bots.parsers.mcafee.parser_atd

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: Parse IoCs from McAfee Advanced Threat Defense reports (hash, IP, URL)

Configuration Parameters

  • Feed parameters (see above)

  • verdict_severity: min report severity to parse

Microsoft CTIP Parser

  • name: intelmq.bots.parsers.microsoft.parser_ctip

  • public: no

  • cache (redis db): none

  • description: Parses data from the Microsoft CTIP Feed

Description

Can parse the JSON format provided by the Interflow interface (lists of dictionaries) as well as the format provided by the Azure interface (one dictionary per line). The provided data differs between the two formats/providers.

The parser is capable of parsing both feeds: - ctip-c2 - ctip-infected-summary The feeds only differ by a few fields, not in the format.

The feeds contain a field called Payload which is nearly always a base64 encoded JSON structure. If decoding works, the contained fields are saved as extra.payload.*, otherwise the field is saved as extra.payload.text.

MISP

  • name: intelmq.bots.parsers.misp.parser

  • public: no

  • cache (redis db): none

  • description: Parses MISP events

Description

MISP events collected by the MISPCollectorBot are passed to this parser for processing. Supported MISP event categories and attribute types are defined in the SUPPORTED_MISP_CATEGORIES and MISP_TYPE_MAPPING class constants.

n6

Information

  • name: intelmq.bots.parsers.n6.parser_n6stomp

  • public: no

  • cache (redis db): none

  • description: Convert n6 data into IntelMQ format.

Configuration Parameters None

Description

Test messages are ignored, this is logged with debug logging level. Also contains a mapping for the classification (results in taxonomy, type and identifier). The name field is normally used as malware.name, if that fails due to disallowed characters, these characters are removed and the original value is saved as event_description.text. This can happen for names like “further iocs: text with invalid ’ char”.

If an n6 message contains multiple IP addresses, multiple events are generated, resulting in events only differing in the address information.

Twitter

Information

  • name: intelmq.bots.parsers.twitter.parser

  • public: no

  • cache (redis db): none

  • description: Extracts URLs from text, fuzzy, aimed at parsing tweets

Configuration Parameters

  • domain_whitelist: domains to be filtered out

  • substitutions: semicolon delimited list of even length of pairs of substitutions (for example: ‘[.];.;,;.’ substitutes ‘[.]’ for ‘.’ and ‘,’ for ‘.’)

  • classification_type: string with a valid classification type as defined in data format

  • default_scheme: Default scheme for URLs if not given. See also the next section.

Default scheme

The dependency url-normalize changed it’s behavior in version 1.4.0 from using http:// as default scheme to https://. Version 1.4.1 added the possibility to specify it. Thus you can only use the default_scheme parameter with a current version of this library >= 1.4.1, with 1.4.0 you will always get https:// as default scheme and for older versions < 1.4.0 http:// is used.

This does not affect URLs which already include the scheme.

Shadowserver

There are two Shadowserver parsers, one for data in CSV format (intelmq.bots.parsers.shadowserver.parser) and one for data in JSON format (intelmq.bots.parsers.shadowserver.parser_json). The latter was added in IntelMQ 2.3 and is meant to be used together with the Shadowserver API collector.

Information

  • name: intelmq.bots.parsers.shadowserver.parser (for CSV data) or intelmq.bots.parsers.shadowserver.parser_json (for JSON data)

  • public: yes

  • description: Parses different reports from Shadowserver.

Configuration Parameters

  • feedname: Optional, the Name of the feed, see list below for possible values.

  • overwrite: If an existing feed.name should be overwritten.

How this bot works?

There are two possibilities for the bot to determine which feed the data belongs to in order to determine the correct mapping of the columns:

Automatic feed detection

Since IntelMQ version 2.1 the parser can detect the feed based on metadata provided by the collector.

When processing a report, this bot takes extra.file_name from the report and looks in config.py how the report should be parsed.

If this lookup is not possible, and the feed name is not given as parameter, the feed cannot be parsed.

The field extra.file_name has the following structure: %Y-%m-%d-${report_name}[-suffix].csv where suffix can be something like country-geo. For example, some possible filenames are 2019-01-01-scan_http-country-geo.csv or 2019-01-01-scan_tftp.csv. The important part is ${report_name}, between the date and the suffix. Since version 2.1.2 the date in the filename is optional, so filenames like scan_tftp.csv are also detected.

Fixed feed name

If the method above is not possible and for upgraded instances, the feed can be set with the feedname parameter. Feed-names are derived from the subjects of the Shadowserver E-Mails. A list of possible feeds can be found in the table below in the column “feed name”.

Supported reports

These are the supported feed name and their corresponding file name for automatic detection:

feed name

file name

Accessible-ADB

scan_adb

Accessible-AFP

scan_afp

Accessible-ARD

scan_ard

Accessible-Cisco-Smart-Install

cisco_smart_install

Accessible-CoAP

scan_coap

Accessible-CWMP

scan_cwmp

Accessible-MS-RDPEUDP

scan_msrdpeudp

Accessible-FTP

scan_ftp

Accessible-Hadoop

scan_hadoop

Accessible-HTTP

scan_http

Accessible-Radmin

scan_radmin

Accessible-RDP

scan_rdp

Accessible-Rsync

scan_rsync

Accessible-SMB

scan_smb

Accessible-Telnet

scan_telnet

Accessible-Ubiquiti-Discovery-Service

scan_ubiquiti

Accessible-VNC

scan_vnc

Blacklisted-IP (deprecated)

blacklist

Blocklist

blocklist

Compromised-Website

compromised_website

DNS-Open-Resolvers

scan_dns

Honeypot-Amplification-DDoS-Events

event4_honeypot_ddos_amp

Honeypot-Brute-Force-Events

event4_honeypot_brute_force

Honeypot-Darknet

event4_honeypot_darknet

Honeypot-HTTP-Scan

event4_honeypot_http_scan

HTTP-Scanners

hp_http_scan

ICS-Scanners

hp_ics_scan

IP-Spoofer-Events

event4_ip_spoofer

Microsoft-Sinkhole-Events IPv4

event4_microsoft_sinkhole

Microsoft-Sinkhole-Events-HTTP IPv4

event4_microsoft_sinkhole_http

NTP-Monitor

scan_ntpmonitor

NTP-Version

scan_ntp

Open-Chargen

scan_chargen

Open-DB2-Discovery-Service

scan_db2

Open-Elasticsearch

scan_elasticsearch

Open-IPMI

scan_ipmi

Open-IPP

scan_ipp

Open-LDAP

scan_ldap

Open-LDAP-TCP

scan_ldap_tcp

Open-mDNS

scan_mdns

Open-Memcached

scan_memcached

Open-MongoDB

scan_mongodb

Open-MQTT

scan_mqtt

Open-MSSQL

scan_mssql

Open-NATPMP

scan_nat_pmp

Open-NetBIOS-Nameservice

scan_netbios

Open-Netis

netis_router

Open-Portmapper

scan_portmapper

Open-QOTD

scan_qotd

Open-Redis

scan_redis

Open-SNMP

scan_snmp

Open-SSDP

scan_ssdp

Open-TFTP

scan_tftp

Open-XDMCP

scan_xdmcp

Outdated-DNSSEC-Key

outdated_dnssec_key

Outdated-DNSSEC-Key-IPv6

outdated_dnssec_key_v6

Sandbox-URL

cwsandbox_url

Sinkhole-DNS

sinkhole_dns

Sinkhole-Events

event4_sinkhole/event6_sinkhole

Sinkhole-Events IPv4

event4_sinkhole

Sinkhole-Events IPv6

event6_sinkhole

Sinkhole-HTTP-Events

event4_sinkhole_http/event6_sinkhole_http

Sinkhole-HTTP-Events IPv4

event4_sinkhole_http

Sinkhole-HTTP-Events IPv6

event6_sinkhole_http

Sinkhole-Events-HTTP-Referer

event4_sinkhole_http_referer/event6_sinkhole_http_referer

Sinkhole-Events-HTTP-Referer IPv4

event4_sinkhole_http_referer

Sinkhole-Events-HTTP-Referer IPv6

event6_sinkhole_http_referer

Spam-URL

spam_url

SSL-FREAK-Vulnerable-Servers

scan_ssl_freak

SSL-POODLE-Vulnerable-Servers

scan_ssl_poodle

Vulnerable-Exchange-Server *

scan_exchange

Vulnerable-ISAKMP

scan_isakmp

Vulnerable-HTTP

scan_http

Vulnerable-SMTP

scan_smtp_vulnerable

* This report can also contain data on active webshells (column tag is exchange;webshell), and are therefore not only vulnerable but also actively infected.

In addition, the following legacy reports are supported:

feed name

successor feed name

file name

Amplification-DDoS-Victim

Honeypot-Amplification-DDoS-Events

ddos_amplification

CAIDA-IP-Spoofer

IP-Spoofer-Events

caida_ip_spoofer

Darknet

Honeypot-Darknet

darknet

Drone

Sinkhole-Events

botnet_drone

Drone-Brute-Force

Honeypot-Brute-Force-Events, Sinkhole-HTTP-Events

drone_brute_force

Microsoft-Sinkhole

Sinkhole-HTTP-Events

microsoft_sinkhole

Sinkhole-HTTP-Drone

Sinkhole-HTTP-Events

sinkhole_http_drone

IPv6-Sinkhole-HTTP-Drone

Sinkhole-HTTP-Events

sinkhole6_http

More information on these legacy reports can be found in Changes in Sinkhole and Honeypot Report Types and Formats.

Development

Structure of this Parser Bot

The parser consists of two files:
  • _config.py

  • parser.py or parser_json.py

Both files are required for the parser to work properly.

Add new Feedformats

Add a new feed format and conversions if required to the file _config.py. Don’t forget to update the mapping dict. It is required to look up the correct configuration.

Look at the documentation in the bot’s _config.py file for more information.

Shodan

Information

  • name: intelmq.bots.parsers.shodan.parser

  • public: yes

  • description: Parses data from Shodan (search, stream etc).

The parser is by far not complete as there are a lot of fields in a big nested structure. There is a minimal mode available which only parses the important/most useful fields and also saves everything in extra.shodan keeping the original structure. When not using the minimal mode if may be useful to ignore errors as many parsing errors can happen with the incomplete mapping.

Configuration Parameters

  • ignore_errors: Boolean (default true)

  • minimal_mode: Boolean (default false)

ZoneH

Information

  • name: intelmq.bots.parsers.zoneh.parser

  • public: yes

  • description: Parses data from ZoneH.

Description This bot is designed to consume defacement reports from zone-h.org. It expects fields normally present in CSV files distributed by email.

Expert Bots

Abusix

Information

  • name: intelmq.bots.experts.abusix.expert

  • lookup: dns

  • public: yes

  • cache (redis db): 5

  • description: RIPE abuse contacts resolving through DNS TXT queries

  • notes: https://abusix.com/contactdb.html

Configuration Parameters

Requirements

This bot can optionally use the python module querycontacts by Abusix itself: https://pypi.org/project/querycontacts/

pip3 install querycontacts

If the package is not installed, our own routines are used.

Aggregate

Information

  • name: intelmq.bots.experts.aggregate.expert

  • lookup: no

  • public: yes

  • cache (redis db): 8

  • description: Aggregates events based upon given fields & timespan

Configuration Parameters

  • Cache parameters (see in section Common parameters)

    • TTL is not used, using it would result in data loss.

  • fields Given fields which are used to aggregate like classification.type, classification.identifier

  • threshold If the aggregated event is lower than the given threshold after the timespan, the event will get dropped.

  • timespan Timespan to aggregate events during the given time. I. e. 1 hour

Usage

Define specific fields to filter incoming events and aggregate them. Also set the timespan you want the events to get aggregated. Usage i. e. 1 hour

Note

The “cleanup” procedure, sends out the aggregated events or drops them based upon the given threshold value. It is called on every incoming message and on the bot’s initialization. If you’re potentially running on low traffic ( no incoming events within the given timestamp ) it is recommended to reload or restart the bot via cronjob each 30 minutes (adapt to your configured timespan). Otherwise you might loose information.

  1. e.:

crontab -e

0,30 * * * *   intelmqctl reload my-aggregate-bot

For reloading/restarting please check the intelmqctl documentation documentation.

ASN Lookup

Information

  • name: intelmq.bots.experts.asn_lookup.expert

  • lookup: local database

  • public: yes

  • cache (redis db): none

  • description: IP to ASN

Configuration Parameters

  • database: Path to the downloaded database.

Requirements

Install pyasn module

pip3 install pyasn

Database

Use this command to create/update the database and reload the bot:

intelmq.bots.experts.asn_lookup.expert --update-database

The database is fetched from routeviews.org and licensed under the Creative Commons Attribution 4.0 International license (see the routeviews FAQ).

CSV Converter

Information

  • name: intelmq.bots.experts.csv_converter.expert

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: Converts an event to CSV format, saved in the output field.

Configuration Parameters

  • delimiter: String, default “,”

  • fieldnames: Comma-separated list of field names, e.g. “time.source,classification.type,source.ip”

Usage

To use the CSV-converted data in an output bot - for example in a file output, use the configuration parameter single_key of the output bot and set it to output.

Cymru Whois

Information

  • name: intelmq.bots.experts.cymru_whois.expert

  • lookup: Cymru DNS

  • public: yes

  • cache (redis db): 5

  • description: IP to geolocation, ASN, BGP prefix

Public documentation: https://www.team-cymru.com/IP-ASN-mapping.html#dns

Configuration Parameters

  • Cache parameters (see in section Common parameters)

  • overwrite: Overwrite existing fields. Default: True if not given (for backwards compatibility, will change in version 3.0.0)

Domain Suffix

This bots adds the public suffix to the event, derived by a domain. See or information on the public suffix list: https://publicsuffix.org/list/ Only rules for ICANN domains are processed. The list can (and should) contain Unicode data, punycode conversion is done during reading.

Note that the public suffix is not the same as the top level domain (TLD). E.g. co.uk is a public suffix, but the TLD is uk. Privately registered suffixes (such as blogspot.co.at) which are part of the public suffix list too, are ignored.

Information

  • name: intelmq.bots.experts.domain_suffix.expert

  • lookup: no

  • public: yes

  • cache (redis db): -

  • description: extracts the domain suffix from the FQDN

Configuration Parameters

  • field: either “fqdn” or “reverse_dns”

  • suffix_file: path to the suffix file

Rule processing

A short summary how the rules are processed:

The simple ones:

com
at
gv.at

example.com leads to com, example.gv.at leads to gv.at.

Wildcards:

*.example.com

www.example.com leads to www.example.com.

And additionally the exceptions, together with the above wildcard rule:

!www.example.com

www.example.com does now not lead to www.example.com, but to example.com.

Database

Use this command to create/update the database and reload the bot:

intelmq.bots.experts.domain_suffix.expert --update-database

Deduplicator

Information

  • name: intelmq.bots.experts.deduplicator.expert

  • lookup: redis cache

  • public: yes

  • cache (redis db): 6

  • description: Bot responsible for ignore duplicated messages. The bot can be configured to perform deduplication just looking to specific fields on the message.

Configuration Parameters

  • Cache parameters (see in section Common parameters)

  • bypass- true or false value to bypass the deduplicator. When set to true, messages will not be deduplicated. Default: false

Parameters for “fine-grained” deduplication

  • filter_type: type of the filtering which can be “blacklist” or “whitelist”. The filter type will be used to define how Deduplicator bot will interpret the parameter filter_keys in order to decide whether an event has already been seen or not, i.e., duplicated event or a completely new event.

    • “whitelist” configuration: only the keys listed in filter_keys will be considered to verify if an event is duplicated or not.

    • “blacklist” configuration: all keys except those in filter_keys will be considered to verify if an event is duplicated or not.

  • filter_keys: string with multiple keys separated by comma. Please note that time.observation key will not be considered even if defined, because the system always ignore that key.

When using a whitelist field pattern and a small number of fields (keys), it becomes more important, that these fields exist in the events themselves. If a field does not exist, but is part of the hashing/deduplication, this field will be ignored. If such events should not get deduplicated, you need to filter them out before the deduplication process, e.g. using a sieve expert. See also this discussion thread on the mailing-list.

Parameters Configuration Example

Example 1

The bot with this configuration will detect duplication only based on source.ip and destination.ip keys.

parameters:
  redis_cache_db: 6
  redis_cache_host: "127.0.0.1"
  redis_cache_password: null
  redis_cache_port: 6379
  redis_cache_ttl: 86400
  filter_type: "whitelist"
  filter_keys: "source.ip,destination.ip"

Example 2

The bot with this configuration will detect duplication based on all keys, except source.ip and destination.ip keys.

parameters:
  redis_cache_db: 6
  redis_cache_host: "127.0.0.1"
  redis_cache_password: null
  redis_cache_port: 6379
  redis_cache_ttl: 86400
  filter_type: "blacklist"
  filter_keys: "source.ip,destination.ip"

Flushing the cache

To flush the deduplicator’s cache, you can use the redis-cli tool. Enter the database used by the bot and submit the flushdb command:

redis-cli -n 6
flushdb

DO Portal Expert Bot

Information

  • name: intelmq.bots.experts.do_portal.expert

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: The DO portal retrieves the contact information from a DO portal instance: http://github.com/certat/do-portal/

Configuration Parameters

  • mode - Either replace or append the new abuse contacts in case there are existing ones.

  • portal_url - The URL to the portal, without the API-path. The used URL is $portal_url + ‘/api/1.0/ripe/contact?cidr=%s’.

  • portal_api_key - The API key of the user to be used. Must have sufficient privileges.

Field Reducer Bot

Information

  • name: intelmq.bots.experts.field_reducer.expert

  • lookup: none

  • public: yes

  • cache (redis db): none

  • description: The field reducer bot is capable of removing fields from events.

Configuration Parameters

  • type - either “whitelist” or “blacklist”

  • keys - Can be a JSON-list of field names ([“raw”, “source.account”]) or a string with a comma-separated list of field names (“raw,source.account”).

Whitelist

Only the fields in keys will passed along.

Blacklist

The fields in keys will be removed from events.

Filter

The filter bot is capable of filtering specific events.

Information

  • name: intelmq.bots.experts.filter.expert

  • lookup: none

  • public: yes

  • cache (redis db): none

  • description: A simple filter for messages (drop or pass) based on a exact string comparison or regular expression

Configuration Parameters

Parameters for filtering with key/value attributes

  • filter_key - key from data format

  • filter_value - value for the key

  • filter_action - action when a message match to the criteria (possible actions: keep/drop)

  • filter_regex - attribute determines if the filter_value shall be treated as regular expression or not.

    If this attribute is not empty (can be true, yes or whatever), the bot uses python’s `re.search <https://docs.python.org/3/library/re.html#re.search>`_ function to evaluate the filter with regular expressions. If this attribute is empty or evaluates to false, an exact string comparison is performed. A check on string inequality can be achieved with the usage of Paths described below.

Parameters for time based filtering

  • not_before - events before this time will be dropped

  • not_after - events after this time will be dropped

Both parameters accept string values describing absolute or relative time:

  • absolute

  • basically anything parseable by datetime parser, eg. “2015-09-012T06:22:11+00:00”

  • time.source taken from the event will be compared to this value to decide the filter behavior

  • relative

  • accepted string formatted like this “<integer> <epoch>”, where epoch could be any of following strings (could optionally end with trailing ‘s’): hour, day, week, month, year

  • time.source taken from the event will be compared to the value (now - relative) to decide the filter behavior

Examples of time filter definition

  • `"not_before" : "2015-09-012T06:22:11+00:00"` events older than the specified time will be dropped

  • `"not_after" : "6 months"` just events older than 6 months will be passed through the pipeline

Possible paths

  • _default: default path, according to the configuration

  • action_other: Negation of the default path

  • filter_match: For all events the filter matched on

  • filter_no_match: For all events the filter does not match

action

match

_default

action_other

filter_match

filter_no_match

keep

keep

drop

drop

In DEBUG logging level, one can see that the message is sent to both matching paths, also if one of the paths is not configured. Of course the message is only delivered to the configured paths.

Format Field

Information

  • name: intelmq.bots.experts.format_field.expert

  • lookup: none

  • cache (redis db): none

  • description: String method operations on column values

Configuration Parameters

Parameters for stripping chars

  • strip_columns - A list of strings or a string of comma-separated values with field names. The names must match the IntelMQ Data Format field names. E.g.

    "columns": [
         "malware.name",
         "extra.tags"
    ],
    

    is equivalent to:

    "columns": "malware.name,extra.tags"
    
  • strip_chars - a set of characters to remove as leading/trailing characters(default: space)

Parameters for replacing chars

  • replace_column - key from data format

  • old_value - the string to search for

  • new_value - the string to replace the old value with

  • replace_count - number specifying how many occurrences of the old value you want to replace(default: 1)

Parameters for splitting string to list of string

  • split_column - key from data format

  • split_separator - specifies the separator to use when splitting the string(default: ,)

Order of operation: strip -> replace -> split. These three methods can be combined such as first strip and then split.

Generic DB Lookup

This bot is capable for enriching intelmq events by lookups to a database. Currently only PostgreSQL and SQLite are supported.

If more than one result is returned, a ValueError is raised.

Information

  • name: intelmq.bots.experts.generic_db_lookup.expert

  • lookup: database

  • public: yes

  • cache (redis db): none

  • description: This bot is capable for enriching intelmq events by lookups to a database.

Configuration Parameters

Connection

  • engine: postgresql or sqlite

  • database: string, defaults to “intelmq”, database name or the SQLite filename

  • table: defaults to “contacts”

PostgreSQL specific

  • host: string, defaults to “localhost”

  • password: string

  • port: integer, defaults to 5432

  • sslmode: string, defaults to “require”

  • user: defaults to “intelmq”

Lookup

  • match_fields: defaults to {“source.asn”: “asn”}

The value is a key-value mapping an arbitrary number intelmq field names to table column names. The values are compared with = only.

Replace fields

  • overwrite: defaults to false. Is applied per field

  • replace_fields: defaults to {“contact”: “source.abuse_contact”}

replace_fields is again a key-value mapping an arbitrary number of table column names to intelmq field names

Gethostbyname

Information

  • name: intelmq.bots.experts.gethostbyname.expert

  • lookup: DNS

  • public: yes

  • cache (redis db): none

  • description: DNS name (FQDN) to IP

Configuration Parameters

  • fallback_to_url If True and no source.fqdn present, use source.url instead while producing source.ip

  • gaierrors_to_ignore: Optional, list (comma-separated) of gaierror codes to ignore, e.g. -3 for EAI_AGAIN (Temporary failure in name resolution). Only accepts the integer values, not the names.

  • overwrite: Boolean. If true, overwrite existing IP addresses. Default: False.

Description

Resolves the source/destination.fqdn hostname using the gethostbyname syscall and saves the resulting IP address as source/destination.ip. The following gaierror resolution errors are ignored and treated as if the hostname cannot be resolved:

  • -2/EAI_NONAME: NAME or SERVICE is unknown

  • -4/EAI_FAIL: Non-recoverable failure in name res.

  • -5/EAI_NODATA: No address associated with NAME.

  • -8/EAI_SERVICE: SERVICE not supported for `ai_socktype’.

  • -11/EAI_SYSTEM: System error returned in `errno’.

Other errors result in an exception if not ignored by the parameter gaierrors_to_ignore (see above). All gaierrors can be found here: http://www.castaglia.org/proftpd/doc/devel-guide/src/lib/glibc-gai_strerror.c.html

HTTP Status

Fetches the HTTP Status for a given URI

Information

  • name: intelmq.bots.experts.http.expert_status

  • description: The bot fetches the HTTP status for a given URL and saves it in the event.

Configuration Parameters

  • field: The name of the field containing the URL to be checked (required).

  • success_status_codes: A list of success status codes. If this parameter is omitted or the list is empty, successful status codes are the ones between 200 and 400.

  • overwrite: Specifies if an existing ‘status’ value should be overwritten.

HTTP Content

Fetches an HTTP resource and checks if it contains a specific string.

Information

  • name: intelmq.bots.experts.http.expert_content

  • description: The bot fetches an HTTP resource and checks if it contains a specific string.

Configuration Parameters

  • field: The name of the field containing the URL to be checked (defaults to source.url)

  • needle: The string that the content available on URL is checked for

  • overwrite: A boolean value that specifies if an existing ‘status’ value should be overwritten.

IDEA Converter

Converts the event to IDEA format and saves it as JSON in the field output. All other fields are not modified.

Documentation about IDEA: https://idea.cesnet.cz/en/index

Information

  • name: intelmq.bots.experts.idea.expert

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: The bot does a best effort translation of events into the IDEA format.

Configuration Parameters

  • test_mode: add Test category to mark all outgoing IDEA events as informal (meant to simplify setting up and debugging new IDEA producers) (default: true)

Lookyloo

Lookyloo is a website screenshotting and analysis tool. For more information and installation instructions visit https://www.lookyloo.eu/

The bot sends a request for source.url to the configured Lookyloo instance and saves the retrieved website screenshot link in the field screenshot_url. Lookyloo only queues the website for screenshotting, therefore the screenshot may not be directly ready after the bot requested it. The pylookyloo library is required for this bot. The http_user_agent parameter is passed on, but not other HTTP-related parameter like proxies.

Events without source.url are ignored.

Information

  • name: intelmq.bots.experts.lookyloo.expert

  • description: LookyLoo expert bot for automated website screenshots

Configuration Parameters

  • instance_url: LookyLoo instance to connect to

MaxMind GeoIP

Information

  • name: intelmq.bots.experts.maxmind_geoip.expert

  • lookup: local database

  • public: yes

  • cache (redis db): none

  • description: IP to geolocation

Setup

The bot requires the MaxMind’s geoip2 Python library, version 2.2.0 has been tested.

To download the database a free license key is required. More information can be found at https://blog.maxmind.com/2019/12/18/significant-changes-to-accessing-and-using-geolite2-databases/

Configuration Parameters

  • database: Path to the local database, e.g. “/opt/intelmq/var/lib/bots/maxmind_geoip/GeoLite2-City.mmdb”

  • overwrite: boolean

  • use_registered: boolean. MaxMind has two country ISO codes: One for the physical location of the address and one for the registered location. Default is false (backwards-compatibility). See also https://github.com/certtools/intelmq/pull/1344 for a short explanation.

  • license_key: License key is necessary for downloading the GeoLite2 database.

Database

Use this command to create/update the database and reload the bot:

intelmq.bots.experts.maxmind_geoip.expert --update-database

MISP

Queries a MISP instance for the source.ip and adds the MISP Attribute UUID and MISP Event ID of the newest attribute found.

Information

  • name: intelmq.bots.experts.misp.expert

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: IP address to MISP attribute and event

Configuration Parameters

  • misp_key: MISP Authkey

  • misp_url: URL of MISP server (with trailing ‘/’)

Generic parameters used in this bot:

  • http_verify_cert: Verify the TLS certificate of the server, boolean (default: true)

McAfee Active Response Hash lookup

Information

  • name: intelmq.bots.experts.mcafee.expert_mar

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: Queries occurrences of hashes within local environment

Configuration Parameters

  • Feed parameters (see above)

  • dxl_config_file: location of file containing required information to connect to DXL bus

  • lookup_type: One of: - Hash: looks up malware.hash.md5, malware.hash.sha1 and malware.hash.sha256 - DestSocket: looks up destination.ip and destination.port - DestIP: looks up destination.ip - DestFQDN: looks up in destination.fqdn

McAfee Active Response lookup

Information

  • name: intelmq.bots.experts.mcafee.expert_mar

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: Queries DXL bus for hashes, IP addresses or FQDNs.

Configuration Parameters

  • Feed parameters (see above)

  • dxl_config_file: location of file containing required information to connect to DXL bus

  • lookup_type: One of <Hash|DestSocket|DestIP|DestFQDN>

Modify

Information

  • name: intelmq.bots.experts.modify.expert

  • lookup: local config

  • public: yes

  • cache (redis db): none

  • description: modify expert bot allows you to change arbitrary field values of events just using a configuration file

Configuration Parameters

  • configuration_path: filename

  • case_sensitive: boolean, default: true

  • maximum_matches: Maximum number of matches. Processing stops after the limit is reached. Default: no limit (null, 0).

  • overwrite: Overwrite any existing fields by matching rules. Default if the parameter is given: true, for backwards compatibility. Default will change to false in version 3.0.0.

Configuration File

The modify expert bot allows you to change arbitrary field values of events just using a configuration file. Thus it is possible to adapt certain values or adding new ones only by changing JSON-files without touching the code of many other bots.

The configuration is called modify.conf and looks like this:

[
    {
        "rulename": "Standard Protocols http",
        "if": {
            "source.port": "^(80|443)$"
        },
        "then": {
            "protocol.application": "http"
        }
    },
    {
        "rulename": "Spamhaus Cert conficker",
        "if": {
            "malware.name": "^conficker(ab)?$"
        },
        "then": {
            "classification.identifier": "conficker"
        }
    },
    {
        "rulename": "bitdefender",
        "if": {
            "malware.name": "bitdefender-(.*)$"
        },
        "then": {
            "malware.name": "{matches[malware.name][1]}"
        }
    },
    {
        "rulename": "urlzone",
        "if": {
            "malware.name": "^urlzone2?$"
        },
        "then": {
            "classification.identifier": "urlzone"
        }
    },
    {
        "rulename": "default",
        "if": {
            "feed.name": "^Spamhaus Cert$"
        },
        "then": {
            "classification.identifier": "{msg[malware.name]}"
        }
    }
]

In our example above we have five groups labeled Standard Protocols http, Spamhaus Cert conficker, bitdefender, urlzone and default. All sections will be considered, in the given order (from top to bottom).

Each rule consists of conditions and actions. Conditions and actions are dictionaries holding the field names of events and regular expressions to match values (selection) or set values (action). All matching rules will be applied in the given order. The actions are only performed if all selections apply.

If the value for a condition is an empty string, the bot checks if the field does not exist. This is useful to apply default values for empty fields.

Actions

You can set the value of the field to a string literal or number.

In addition you can use the standard Python string format syntax to access the values from the processed event as msg and the match groups of the conditions as matches, see the bitdefender example above. Group 0 ([0]) contains the full matching string. See also the documentation on re.Match.group.

Note that matches will also contain the match groups from the default conditions if there were any.

Examples

We have an event with feed.name = Spamhaus Cert and malware.name = confickerab. The expert loops over all sections in the file and eventually enters section Spamhaus Cert. First, the default condition is checked, it matches! OK, going on. Otherwise the expert would have selected a different section that has not yet been considered. Now, go through the rules, until we hit the rule conficker. We combine the conditions of this rule with the default conditions, and both rules match! So we can apply the action: classification.identifier is set to conficker, the trivial name.

Assume we have an event with feed.name = Spamhaus Cert and malware.name = feodo. The default condition matches, but no others. So the default action is applied. The value for classification.identifier will be set to feodo by {msg[malware.name]}.

Types

If the rule is a string, a regular expression search is performed, also for numeric values (str() is called on them). If the rule is numeric for numeric values, a simple comparison is done. If other types are mixed, a warning will be thrown.

For boolean values, the comparison value needs to be true or false as in JSON they are written all-lowercase.

National CERT contact lookup by CERT.AT

Information

  • name: intelmq.bots.experts.national_cert_contact_certat.expert

  • lookup: https

  • public: yes

  • cache (redis db): none

  • description: https://contacts.cert.at offers an IP address to national CERT contact (and cc) mapping. See https://contacts.cert.at for more info.

Configuration Parameters

  • filter: (true/false) act as a filter for AT.

  • overwrite_cc: set to true if you want to overwrite any potentially existing cc fields in the event.

RDAP

Information

  • name: intelmq.bots.experts.rdap.expert

  • lookup: http/https

  • public: yes/no

  • cache (redis db): 5

  • description: Asks rdap servers for a given domain.

Configuration Parameters

  • rdap_order: a list of strings, default ['abuse', 'technical']. Search order of contacts with these roles.

  • rdap_bootstrapped_servers: Customized RDAP servers. Do not forget the trailing slash. For example:

{
   "at": {
      "url": "rdap.server.at/v1/,
      "auth": {
         "type": "jwt",
         "token": "ey..."
      }
   },
   "de": "rdap.service:1337/v1/"
}

RecordedFuture IP risk

This Bot tags events with score found in recorded futures large IP risklist.

Information

  • name: intelmq.bots.experts.recordedfuture_iprisk.expert

  • lookup: local database

  • public: no

  • cache (redis db): none

  • description: Record risk score associated to source and destination IP if they are present. Assigns 0 to IP addresses not in the RF list.

Configuration Parameters

  • database: Location of csv file obtained from recorded future API (a script is provided to download the large IP set)

  • overwrite: set to true if you want to overwrite any potentially existing risk score fields in the event.

  • api_token: This needs to contain valid API token to download the latest database data.

Description

For both source.ip and destination.ip the corresponding risk score is fetched from a local database created from Recorded Future’s API. The score is recorded in extra.rf_iprisk.source and extra.rf_iprisk.destination. If a lookup for an IP fails a score of 0 is recorded.

See https://www.recordedfuture.com/products/api/ and speak with your recorded future representative for more information.

The list is obtained from recorded future API and needs a valid API TOKEN The large list contains all IP’s with a risk score of 25 or more. If IP’s are not present in the database a risk score of 0 is given

A script is supplied that may be run as intelmq to update the database.

Database

Use this command to create/update the database and reload the bot:

intelmq.bots.experts.recordedfuture_iprisk.expert --update-database

Reverse DNS

For both source.ip and destination.ip the PTR record is fetched and the first valid result is used for source.reverse_dns/destination.reverse_dns.

Information

  • name: intelmq.bots.experts.reverse_dns.expert

  • lookup: DNS

  • public: yes

  • cache (redis db): 8

  • description: IP to domain

Configuration Parameters

  • Cache parameters (see in section Common parameters)

  • cache_ttl_invalid_response: The TTL for cached invalid responses.

  • overwrite: Overwrite existing fields. Default: True if not given (for backwards compatibility, will change in version 3.0.0)

RFC1918

Several RFCs define ASNs, IP Addresses and Hostnames (and TLDs) reserved for documentation. Events or fields of events can be dropped if they match the criteria of either being reserved for documentation (e.g. AS 64496, Domain example.com) or belonging to a local area network (e.g. 192.168.0.0/24). These checks can applied to URLs, IP Addresses, FQDNs and ASNs.

It is configurable if the whole event should be dropped (“policies”) or just the field removed, as well as which fields should be checked.

Sources:

Information

  • name: intelmq.bots.experts.rfc1918.expert

  • lookup: none

  • public: yes

  • cache (redis db): none

  • description: removes events or single fields with invalid data

Configuration Parameters

  • fields: string, comma-separated list of fields e.g. destination.ip,source.asn,source.url. Supported fields are:

    • destination.asn & source.asn

    • destination.fqdn & source.fqdn

    • destination.ip & source.ip

    • destination.url & source.url

  • policy: string, comma-separated list of policies, e.g. del,drop,drop. drop will cause that the the entire event to be removed if the field is , del causes the field to be removed.

With the example parameter values given above, this means that:

  • If a destination.ip value is part of a reserved network block, the field will be removed (policy “del”).

  • If a source.asn value is in the range of reserved AS numbers, the event will be removed altogether (policy “drop).

  • If a source.url value contains a host with either an IP address part of a reserved network block, or a reserved domain name (or with a reserved TLD), the event will be dropped (policy “drop”)

RIPE

Online RIPE Abuse Contact and Geolocation Finder for IP addresses and Autonomous Systems.

Information

  • name: intelmq.bots.experts.ripe.expert

  • lookup: HTTPS API

  • public: yes

  • cache (redis db): 10

  • description: IP to abuse contact

Configuration Parameters

  • Cache parameters (see section Common parameters)

  • mode: either append (default) or replace

  • query_ripe_db_asn: Query for IPs at http://rest.db.ripe.net/abuse-contact/%s.json, default true

  • query_ripe_db_ip: Query for ASNs at http://rest.db.ripe.net/abuse-contact/as%s.json, default true

  • query_ripe_stat_asn: Query for ASNs at https://stat.ripe.net/data/abuse-contact-finder/data.json?resource=%s, default true

  • query_ripe_stat_ip: Query for IPs at https://stat.ripe.net/data/abuse-contact-finder/data.json?resource=%s, default true

  • query_ripe_stat_geolocation: Query for IPs at https://stat.ripe.net/data/maxmind-geo-lite/data.json?resource=%s, default true

Sieve

Information

  • name: intelmq.bots.experts.sieve.expert

  • lookup: none

  • public: yes

  • cache (redis db): none

  • description: Filtering with a sieve-based configuration language

Configuration Parameters

  • file: Path to sieve file. Syntax can be validated with intelmq_sieve_expert_validator.

Description

The sieve bot is used to filter and/or modify events based on a set of rules. The rules are specified in an external configuration file and with a syntax similar to the Sieve language used for mail filtering.

Each rule defines a set of matching conditions on received events. Events can be matched based on keys and values in the event. Conditions can be combined using parenthesis and the boolean operators && and ||. If the processed event matches a rule’s conditions, the corresponding actions are performed. Actions can specify whether the event should be kept or dropped in the pipeline (filtering actions) or if keys and values should be changed (modification actions).

Requirements

To use this bot, you need to install the required dependencies:

pip3 install -r intelmq/bots/experts/sieve/REQUIREMENTS.txt

Examples

The following excerpts illustrate some of the basic features of the sieve file format:

if :exists source.fqdn {
  keep  // aborts processing of subsequent rules and forwards the event.
}


if :notexists source.abuse_contact || source.abuse_contact =~ '.*@example.com' {
  drop  // aborts processing of subsequent rules and drops the event.
}

if source.ip << '192.0.0.0/24' {
    add! comment = 'bogon' // sets the field comment to this value and overwrites existing values
    path 'other-path' // the message is sent to the given path
}

if classification.type == ['phishing', 'malware-distribution'] && source.fqdn =~ '.*\.(ch|li)$' {
  add! comment = 'domainabuse'
  keep
} elif classification.type == 'scanner' {
  add! comment = 'ignore'
  drop
} else {
  remove comment
}

Reference

Sieve File Structure

The sieve file contains an arbitrary number of rules of the form:

if EXPRESSION {
    ACTIONS
} elif EXPRESSION {
    ACTIONS
} else {
    ACTIONS
}

Nested if-statements and mixed if statements and rules in the same scope are possible.

Expressions

Each rule specifies on or more expressions to match an event based on its keys and values. Event keys are specified as strings without quotes. String values must be enclosed in single quotes. Numeric values can be specified as integers or floats and are unquoted. IP addresses and network ranges (IPv4 and IPv6) are specified with quotes. List values for use with list/set operators are specified as string, float, int, bool and string literals separated by commas and enclosed in square brackets. Expression statements can be combined and chained using parentheses and the boolean operators && and ||. The following operators may be used to match events:

  • :exists and :notexists match if a given key exists, for example:

    if :exists source.fqdn { ... }

  • == and != match for equality of strings, numbers, and booleans, for example:

    if feed.name != 'acme-security' || feed.accuracy == 100 || extra.false_positive == false { ... }

  • :contains matches on substrings.

  • =~ matches strings based on the given regular expression. !~ is the inverse regular expression match.

  • Numerical comparisons are evaluated with <, <=, >, >=.

  • << matches if an IP address is contained in the specified network range:

    if source.ip << '10.0.0.0/8' { ... }

  • String values to match against can also be specified as lists of strings, which have separate operators. For example:

    if source.ip :in ['8.8.8.8', '8.8.4.4'] { ... }

In this case, the event will match if it contains a key source.ip with either value 8.8.8.8 or 8.8.4.4.

There are also :containsany to match at least one of a list of substrings, and :regexin to match at least one of a list of regular expressions, similar to the :contains and =~ operators.

  • Lists of numeric values support :in to check for inclusion in a list of numbers:

    if source.port :in [80, 443] { ... }

  • :equals tests for equality between lists, including order. Example for checking a hostname-port pair: if extra.host_tuple :equals ['dns.google', 53] { ... }

  • :setequals tests for set-based equality (ignoring duplicates and value order) between a list of given values. Example for checking for the first nameserver of two domains, regardless of the order they are given in the list: if extra.hostnames :setequals ['ns1.example.com', 'ns1.example.mx'] { ... }

  • :overlaps tests if there is at least one element in common between the list specified by a key and a list of values. Example for checking if at least one of the ICS, database or vulnerable tags is given: ``if extra.tags :overlaps [‘ics’, ‘database’, ‘vulnerable’] { … } ``

  • :subsetof tests if the list of values from the given key only contains values from a set of values specified as the argument. Example for checking for a host that has only ns1.example.com and/or ns2.[…] as its apparent hostname: if extra.hostnames :subsetof ['ns1.example.com', 'ns2.example.com'] { ... }

  • :supersetof tests if the list of values from the given key is a superset of the values specified as the argument. Example for matching hosts with at least the IoT and vulnerable tags: if extra.tags :supersetof ['iot', 'vulnerable'] { ... }

  • Boolean values can be matched with == or != followed by true or false. Example: if extra.has_known_vulns == true { ... }

  • The combination of multiple expressions can be done using parenthesis and boolean operators:

if (source.ip == '127.0.0.1') && (comment == 'add field' || classification.taxonomy == 'vulnerable') { ... }

  • Any single expression or a parenthesised group of expressions can be negated using !:

if ! source.ip :contains '127.0.0.' || ! ( source.ip == '172.16.0.5' && source.port == 25 ) { ... }

  • Note: Since 3.0.0, list-based operators are used on list values, such as foo :in [1, 2, 3] instead of foo == [1, 2, 3] and foo :regexin [‘.mx’, ‘.zz’] rather than foo =~ [‘.mx’, ‘.zz’], and similarly for :containsany vs :contains. Besides that, ``:notcontains` has been removed, with e.g foo :notcontains [‘.mx’, ‘.zz’] now being represented using negation as ! foo :contains [‘.mx’, ‘.zz’].

Actions

If part of a rule matches the given conditions, the actions enclosed in { and } are applied. By default, all events that are matched or not matched by rules in the sieve file will be forwarded to the next bot in the pipeline, unless the drop action is applied.

  • add adds a key value pair to the event. It can be a string, number, or boolean. This action only applies if the key is not yet defined in the event. If the key is already defined, the action is ignored. Example:

    add comment = 'hello, world'

    Some basic mathematical expressions are possible, but currently support only relative time specifications objects are supported. For example: `add time.observation += '1 hour'` `add time.observation -= '10 hours'`

  • add! same as above, but will force overwrite the key in the event.

  • update modifies an existing value for a key. Only applies if the key is already defined. If the key is not defined in the event, this action is ignored. This supports mathematical expressions like above. Example:

    update feed.accuracy = 50

    Some basic mathematical expressions are possible, but currently support only relative time specifications objects are supported. For example: `update time.observation += '1 hour'` `update time.observation -= '10 hours'`

  • remove removes a key/value from the event. Action is ignored if the key is not defined in the event. Example:

    remove extra.comments

  • keep sends the message to the next bot in the pipeline (same as the default behaviour), and stops sieve file processing.

    keep

  • path sets the path (named queue) the message should be sent to (implicitly or with the command keep. The named queue needs to configured in the pipeline, see the User Guide for more information.

    path 'named-queue'

    You can as well set multiple destination paths with the same syntax as for value lists:

    path ['one', 'two']

    This will result in two identical message, one sent to the path one and the other sent to the path two.

    If the path is not configured, the error looks like:

    ```
    File “/path/to/intelmq/intelmq/lib/pipeline.py”, line 353, in send

    for destination_queue in self.destination_queues[path]:

    KeyError: ‘one’ ```

  • drop marks the event to be dropped. The event will not be forwarded to the next bot in the pipeline. The sieve file processing is interrupted upon reaching this action. No other actions may be specified besides the drop action within { and }.

Comments

Comments may be used in the sieve file: all characters after // and until the end of the line will be ignored.

Validating a sieve file

Use the following command to validate your sieve files:

$ intelmq.bots.experts.sieve.validator
usage: intelmq.bots.experts.sieve.validator [-h] sievefile

Validates the syntax of sievebot files.

positional arguments:
  sievefile   Sieve file

optional arguments:
  -h, --help  show this help message and exit

Taxonomy

Information

  • name: intelmq.bots.experts.taxonomy.expert

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: Adds the classification.taxonomy field according to the RSIT taxonomy.

Please note that there is a slight mismatch of IntelMQ’s taxonomy to the upstream taxonomy, but it should not matter here much.

Configuration Parameters

None.

Description

Information on the “Reference Security Incident Taxonomy” can be found here: https://github.com/enisaeu/Reference-Security-Incident-Taxonomy-Task-Force

For brevity, “type” means classification.type and “taxonomy” means classification.taxonomy.

  • If taxonomy is missing, and type is given, the according taxonomy is set.

  • If neither taxonomy, not type is given, taxonomy is set to “other” and type to “unknown”.

  • If taxonomy is given, but type is not, type is set to “unknown”.

Threshold

Information

  • Cache parameters (see section Common parameters)

  • name: intelmq.bots.experts.threshold.expert

  • lookup: redis cache

  • public: no

  • cache (redis db): 11

  • description: Check if the number of similar messages during a specified time interval exceeds a set value.

Configuration Parameters

  • filter_keys: String, comma-separated list of field names to consider or ignore when determining which messages are similar.

  • filter_type: String, whitelist (consider only the fields in filter_keys) or blacklist (consider everything but the fields in filter_keys).

  • timeout: Integer, number of seconds before threshold counter is reset.

  • threshold: Integer, number of messages required before propagating one. In forwarded messages, the threshold is saved in the message as extra.count.

  • add_keys: Array of string->string, optional, fields and values to add (or update) to propagated messages. Example:

    "add_keys": {
        "classification.type": "spam",
        "comment": "Started more than 10 SMTP connections"
    }
    

Limitations

This bot has certain limitations and is not a true threshold filter (yet). It works like this:

  1. Every incoming message is hashed according to the filter_* parameters.

  2. The hash is looked up in the cache and the count is incremented by 1, and the TTL of the key is (re-)set to the timeout.

  3. If the new count matches the threshold exactly, the message is forwarded. Otherwise it is dropped.

Please note: Even if a message is sent, any further identical messages are dropped, if the time difference to the last message is less than the timeout! The counter is not reset if the threshold is reached.

Tor Nodes

Information

  • name: intelmq.bots.experts.tor_nodes.expert

  • lookup: local database

  • public: yes

  • cache (redis db): none

  • description: check if IP is tor node

Configuration Parameters

  • database: Path to the database

Database

Use this command to create/update the database and reload the bot:

intelmq.bots.experts.tor_nodes.expert --update-database

Trusted Introducer Lookup Expert

Information

  • name: intelmq.bots.experts.trusted_introducer_lookup.expert

  • lookup: internet

  • public: yes

  • cache (redis db): none

  • description: Lookups data from trusted introducer public teams list.

Configuration Parameters

  • order: Possible values are ‘domain’, ‘asn’. You can set multiple values, so first match wins.

  • If ‘domain’ is set, it will lookup the source.fqdn field. It will go from high-order to low-order, i.e. 1337.super.example.com -> super.example.com -> example.com -> .com

  • If ‘asn’ is set, it will lookup source.asn.

After a match, the abuse contact will be fetched from the trusted introducer teams list and will be stored in the event as source.abuse_contact. If there is no match, the event will not be enriched and will be sent to the next configured step.

Tuency

Information

Configuration Parameters

  • url: Tuency instance URL. Without the API path.

  • authentication_token: The Bearer authentication token. Without the Bearer prefix.

  • overwrite: Boolean, if existing data in source.abuse_contact should be overwritten. Default: true

Description

tuency is a contact management database addressing the needs of CERTs. Users of tuency can configure contact addresses and delivery settings for IP objects (addresses, netblocks), Autonomous Systems, and (sub-)domains. This expert queries the information for source.ip and source.fqdn using the following other fields:

  • classification.taxonomy

  • classification.type

  • feed.provider

  • feed.name

These fields therefore need to exist, otherwise the message is skipped.

The API parameter “feed_status” is currently set to “production” constantly, until IntelMQ supports this field.

The API answer is processed as following. For the notification interval:

  • If suppress is true, then extra.notify is set to false.

  • Otherwise:

    • If the interval is immediate, then extra.ttl is set to 0.

    • Otherwise the interval is converted into seconds and saved in extra.ttl.

For the contact lookup: For both fields ip and domain, the destinations objects are iterated and its email fields concatenated to a comma-separated list in source.abuse_contact.

The IntelMQ fields used by this bot may change in the next IntelMQ release, as soon as better suited fields are available.

Url2FQDN

This bot extracts the Host from the source.url and destination.url fields and writes it to source.fqdn or destination.fqdn if it is a hostname, or source.ip or destination.ip if it is an IP address.

Information

  • name: intelmq.bots.experts.url2fqdn.expert

  • lookup: none

  • public: yes

  • cache (redis db): none

  • description: writes domain name from URL to FQDN or IP address

Configuration Parameters

  • overwrite: boolean, replace existing FQDN / IP address?

uWhoisd

uWhoisd is a universal Whois server that supports caching and stores whois entries for historical purposes.

The bot sends a request for source.url, source.fqdn, source.ip or source.asn to the configured uWhoisd instance and saves the retrieved whois entry:

  • If both source.url and source.fqdn are present, it will only do a request for source.fqdn, as the hostname of source.url should be the same as source.fqdn. The whois entry will be saved in extra.whois.fqdn.

  • If source.ip is present, the whois entry will be saved in extra.whois.ip

  • If source.asn is present, he whois entry will be saved in extra.whois.asn

Events without source.url, source.fqdn, source.ip, or source.asn, are ignored.

Note: requesting a whois entry for a fully qualified domain name (FQDN) only works if the request only contains the domain. uWhoisd will automatically strip the subdomain part if it is present in the request.

Example: https://www.theguardian.co.uk

The whois request will be for theguardian.co.uk

Information

  • name: intelmq.bots.experts.uwhoisd.expert

  • description: uWhoisd is a universal Whois server

Configuration Parameters

  • server: IP or hostname to connect to (default: localhost)

  • port: Port to connect to (default: 4243)

Wait

Information

  • name: intelmq.bots.experts.wait.expert

  • lookup: none

  • public: yes

  • cache (redis db): none

  • description: Waits for a some time or until a queue size is lower than a given number.

Configuration Parameters

  • queue_db: Database number of the database, default 2. Converted to integer.

  • queue_host: Host of the database, default localhost.

  • queue_name: Name of the queue to be watched, default null. This is not the name of a bot but the queue’s name.

  • queue_password: Password for the database, default None.

  • queue_polling_interval: Interval to poll the list length in seconds. Converted to float.

  • queue_port: Port of the database, default 6379. Converted to integer.

  • queue_size: Maximum size of the queue, default 0. Compared by <=. Converted to integer.

  • sleep_time: Time to sleep before sending the event.

Only one of the two modes is possible. If a queue name is given, the queue mode is active. If the sleep_time is a number, sleep mode is active. Otherwise the dummy mode is active, the events are just passed without an additional delay.

Note that SIGHUPs and reloads interrupt the sleeping.

Output Bots

AMQP Topic

Sends data to an AMQP Server See https://www.rabbitmq.com/tutorials/amqp-concepts.html for more details on amqp topic exchange.

Requires the pika python library.

Information

  • name: intelmq.bots.outputs.amqptopic.output

  • lookup: to the amqp server

  • public: yes

  • cache: no

  • description: Sends the event to a specified topic of an AMQP server

Configuration parameters

  • connection_attempts : The number of connection attempts to defined server, defaults to 3

  • connection_heartbeat : Heartbeat to server, in seconds, defaults to 3600

  • connection_host : Name/IP for the AMQP server, defaults to 127.0.0.1

  • connection_port : Port for the AMQP server, defaults to 5672

  • connection_vhost : Virtual host to connect, on an http(s) connection would be http:/IP/<your virtual host>

  • content_type : Content type to deliver to AMQP server, currently only supports “application/json”

  • delivery_mode : 1 - Non-persistent, 2 - Persistent. On persistent mode, messages are delivered to ‘durable’ queues and will be saved to disk.

  • exchange_durable : If set to True, the exchange will survive broker restart, otherwise will be a transient exchange.

  • exchange_name : The name of the exchange to use

  • exchange_type : Type of the exchange, e.g. topic, fanout etc.

  • keep_raw_field : If set to True, the message ‘raw’ field will be sent

  • password : Password for authentication on your AMQP server

  • require_confirmation : If set to True, an exception will be raised if a confirmation error is received

  • routing_key : The routing key for your amqptopic

  • single_key : Only send the field instead of the full event (expecting a field name as string)

  • username : Username for authentication on your AMQP server

  • use_ssl : Use ssl for the connection, make sure to also set the correct port, usually 5671 (true/false)

  • message_hierarchical_output: Convert the message to hierarchical JSON, default: false

  • message_with_type : Include the type in the sent message, default: false

  • message_jsondict_as_string: Convert fields of type JSONDict (extra) as string, default: false

If no authentication should be used, leave username or password empty or null.

Examples of usage

  • Useful to send events to a RabbitMQ exchange topic to be further processed in other platforms.

Confirmation

If routing key or exchange name are invalid or non existent, the message is accepted by the server but we receive no confirmation. If parameter require_confirmation is True and no confirmation is received, an error is raised.

Common errors

Unroutable messages / Undefined destination queue

The destination exchange and queue need to exist beforehand, with your preferred settings (e.g. durable, lazy queue. If the error message says that the message is “unroutable”, the queue doesn’t exist.

Blackhole

This output bot discards all incoming messages.

Information

  • name: intelmq.bots.outputs.blackhole.output

  • lookup: no

  • public: yes

  • cache: no

  • description: discards messages

Elasticsearch Output Bot

Information

  • name: intelmq.bots.outputs.elasticsearch.output

  • lookup: yes

  • public: yes

  • cache: no

  • description: Output Bot that sends events to Elasticsearch

Only ElasticSearch version 7 supported.

It is also possible to feed data into ElasticSearch using ELK-Stack via Redis and Logstash, see ELK Stack for more information. This methods supports various different versions of ElasticSearch.

Configuration parameters

  • elastic_host: Name/IP for the Elasticsearch server, defaults to 127.0.0.1

  • elastic_port: Port for the Elasticsearch server, defaults to 9200

  • elastic_index: Index for the Elasticsearch output, defaults to intelmq

  • rotate_index: If set, will index events using the date information associated with the event.

    Options: ‘never’, ‘daily’, ‘weekly’, ‘monthly’, ‘yearly’. Using ‘intelmq’ as the elastic_index, the following are examples of the generated index names:

    'never' --> intelmq
    'daily' --> intelmq-2018-02-02
    'weekly' --> intelmq-2018-42
    'monthly' --> intelmq-2018-02
    'yearly' --> intelmq-2018
    
  • http_username: HTTP basic authentication username

  • http_password: HTTP basic authentication password

  • use_ssl: Whether to use SSL/TLS when connecting to Elasticsearch. Default: False

  • http_verify_cert: Whether to require verification of the server’s certificate. Default: False

  • ssl_ca_certificate: An optional path to a certificate bundle to use for verifying the server

  • ssl_show_warnings: Whether to show warnings if the server’s certificate cannot be verified. Default: True

  • replacement_char: If set, dots (‘.’) in field names will be replaced with this character prior to indexing. This is for backward compatibility with ES 2.X. Default: null. Recommended for ES2.X: ‘_’

  • flatten_fields: In ES, some query and aggregations work better if the fields are flat and not JSON. Here you can provide a list of fields to convert.

    Can be a list of strings (fieldnames) or a string with field names separated by a comma (,). eg extra,field2 or [‘extra’, ‘field2’] Default: [‘extra’]

See contrib/elasticsearch/elasticmapper for a utility for creating Elasticsearch mappings and templates.

If using rotate_index, the resulting index name will be of the form [elastic_index]-[event date]. To query all intelmq indices at once, use an alias (https://www.elastic.co/guide/en/elasticsearch/reference/current/indices-aliases.html), or a multi-index query.

The data in ES can be retrieved with the HTTP-Interface:

> curl -XGET 'http://localhost:9200/intelmq/events/_search?pretty=True'

File

Information

  • name: intelmq.bots.outputs.file.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: output messages (reports or events) to file

Multihreading is disabled for this bot, as this would lead to corrupted files.

Configuration Parameters

  • encoding_errors_mode: By default ‘strict’, see for more details and options: https://docs.python.org/3/library/functions.html#open For example with ‘backslashreplace’ all characters which cannot be properly encoded will be written escaped with backslashes.

  • file: file path of output file. Missing directories will be created if possible with the mode 755.

  • format_filename: Boolean if the filename should be formatted (default: false).

  • hierarchical_output: If true, the resulting dictionary will be hierarchical (field names split by dot).

  • single_key: if none, the whole event is saved (default); otherwise the bot saves only contents of the specified key. In case of raw the data is base64 decoded.

Filename formatting

The filename can be formatted using pythons string formatting functions if format_filename is set. See https://docs.python.org/3/library/string.html#formatstrings

For example:
  • The filename …/{event[source.abuse_contact]}.txt will be (for example) …/abuse@example.com.txt.

  • …/{event[time.source]:%Y-%m-%d} results in the date of the event used as filename.

If the field used in the format string is not defined, None will be used as fallback.

Files

Information

  • name: intelmq.bots.outputs.files.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: saving of messages as separate files

Configuration Parameters

  • dir: output directory (default /opt/intelmq/var/lib/bots/files-output/incoming)

  • tmp: temporary directory (must reside on the same filesystem as dir) (default: /opt/intelmq/var/lib/bots/files-output/tmp)

  • suffix: extension of created files (default .json)

  • hierarchical_output: if true, use nested dictionaries; if false, use flat structure with dot separated keys (default)

  • single_key: if none, the whole event is saved (default); otherwise the bot saves only contents of the specified key

McAfee Enterprise Security Manager

Information

  • name: intelmq.bots.outputs.mcafee.output_esm_ip

  • lookup: yes

  • public: no

  • cache (redis db): none

  • description: Writes information out to McAfee ESM watchlist

Configuration Parameters

  • Feed parameters (see above)

  • esm_ip: IP address of ESM instance

  • esm_user: username of user entitled to write to watchlist

  • esm_pw: password of user

  • esm_watchlist: name of the watchlist to write to

  • field: name of the IntelMQ field to be written to ESM

MISP Feed

Information

  • name: intelmq.bots.outputs.misp.output_feed

  • lookup: no

  • public: no

  • cache (redis db): none

  • description: Create a directory layout in the MISP Feed format

The PyMISP library >= 2.4.119.1 is required, see REQUIREMENTS.txt.

Configuration Parameters

  • Feed parameters (see above)

  • misp_org_name: Org name which creates the event, string

  • misp_org_uuid: Org UUID which creates the event, string

  • output_dir: Output directory path, e.g. /opt/intelmq/var/lib/bots/mispfeed-output. Will be created if it does not exist and possible.

  • interval_event: The output bot creates one event per each interval, all data in this time frame is part of this event. Default “1 hour”, string.

Usage in MISP

Configure the destination directory of this feed as feed in MISP, either as local location, or served via a web server. See the MISP documentation on Feeds for more information

MISP API

Information

  • name: intelmq.bots.outputs.misp.output_api

  • lookup: no

  • public: no

  • cache (redis db): none

  • description: Connect to a MISP instance and add event as MISPObject if not there already.

The PyMISP library >= 2.4.120 is required, see REQUIREMENTS.txt.

Configuration Parameters

  • Feed parameters (see above)

  • add_feed_provider_as_tag: boolean (use true when in doubt)

  • add_feed_name_as_tag: boolean (use true when in doubt)

  • misp_additional_correlation_fields: list of fields for which the correlation flags will be enabled (in addition to those which are in significant_fields)

  • misp_additional_tags: list of tags to set not be searched for when looking for duplicates

  • misp_key: string, API key for accessing MISP

  • misp_publish: boolean, if a new MISP event should be set to “publish”.

    Expert setting as MISP may really make it “public”! (Use false when in doubt.)

  • misp_tag_for_bot: string, used to mark MISP events

  • misp_to_ids_fields: list of fields for which the to_ids flags will be set

  • misp_url: string, URL of the MISP server

  • significant_fields: list of intelmq field names

The significant_fields values will be searched for in all MISP attribute values and if all values are found in the same MISP event, no new MISP event will be created. Instead if the existing MISP events have the same feed.provider and match closely, their timestamp will be updated.

If a new MISP event is inserted the significant_fields and the misp_additional_correlation_fields will be the attributes where correlation is enabled.

Make sure to build the IntelMQ Botnet in a way the rate of incoming events is what MISP can handle, as IntelMQ can process many more events faster than MISP (which is by design as MISP is for manual handling). Also remove the fields of the IntelMQ events with an expert bot that you do not want to be inserted into MISP.

(More details can be found in the docstring of output_api.py.

MongoDB

Saves events in a MongoDB either as hierarchical structure or flat with full key names. time.observation and time.source are saved as datetime objects, not as ISO formatted string.

Information

  • name: intelmq.bots.outputs.mongodb.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: MongoDB is the bot responsible to send events to a MongoDB database

Configuration Parameters

  • collection: MongoDB collection

  • database: MongoDB database

  • db_user : Database user that should be used if you enabled authentication

  • db_pass : Password associated to db_user

  • host: MongoDB host (FQDN or IP)

  • port: MongoDB port, default: 27017

  • hierarchical_output: Boolean (default true) as MongoDB does not allow saving keys with dots, we split the dictionary in sub-dictionaries.

  • replacement_char: String (default ‘_’) used as replacement character for the dots in key names if hierarchical output is not used.

Installation Requirements

pip3 install pymongo>=2.7.1

The bot has been tested with pymongo versions 2.7.1, 3.4 and 3.10.1 (server versions 2.6.10 and 3.6.8).

Redis

Information

  • name: intelmq.bots.outputs.redis.output

  • lookup: to the Redis server

  • public: yes

  • cache (redis db): none

  • description: Output Bot that sends events to a remote Redis server/queue.

Configuration Parameters

  • redis_db: remote server database, e.g.: 2

  • redis_password: remote server password

  • redis_queue: remote server list (queue), e.g.: “remote-server-queue”

  • redis_server_ip: remote server IP address, e.g.: 127.0.0.1

  • redis_server_port: remote server Port, e.g.: 6379

  • redis_timeout: Connection timeout, in milliseconds, e.g.: 50000

  • hierarchical_output: whether output should be sent in hierarchical JSON format (default: false)

  • with_type: Send the __type field (default: true)

Examples of usage

  • Can be used to send events to be processed in another system. E.g.: send events to Logstash.

  • In a multi tenant installation can be used to send events to external/remote IntelMQ instance. Any expert bot queue can receive the events.

  • In a complex configuration can be used to create logical sets in IntelMQ-Manager.

Request Tracker

Information

  • name: intelmq.bots.outputs.rt.output

  • lookup: to the Request Tracker instance

  • public: yes

  • cache (redis db): none

  • description: Output Bot that creates Request Tracker tickets from events.

Description

The bot creates tickets in Request Tracker and uses event fields for the ticket body text. The bot follows the workflow of the RTIR:

  • create ticket in Incidents queue (or any other queue)

    • all event fields are included in the ticket body,

    • event attributes are assigned to tickets’ CFs according to the attribute mapping,

    • ticket taxonomy can be assigned according to the CF mapping. If you use taxonomy different from ENISA RSIT, consider using some extra attribute field and do value mapping with modify or sieve bot,

  • create linked ticket in Investigations queue, if these conditions are met

    • if first ticket destination was Incidents queue,

    • if there is source.abuse_contact is specified,

    • if description text is specified in the field appointed by configuration,

  • RT/RTIR supposed to do relevant notifications by scrip working on condition “On Create”,

  • configuration option investigation_fields specifies which event fields has to be included in the investigation,

  • Resolve Incident ticket, according to configuration (Investigation ticket status should depend on RT scrip configuration),

Take extra caution not to flood your ticketing system with enormous amount of tickets. Add extra filtering for that to pass only critical events to the RT, and/or deduplicating events.

Configuration Parameters

  • rt_uri, rt_user, rt_password, verify_cert: RT API endpoint connection details, string.

  • queue: ticket destination queue. If set to ‘Incidents’, ‘Investigations’ ticket will be created if create_investigation is set to true, string.

  • CF_mapping: mapping attributes to ticket CFs, dictionary. E.g {“event_description.text”:”Description”,”source.ip”:”IP”,”extra.classification.type”:”Incident Type”,”classification.taxonomy”:”Classification”}

  • final_status: the final status for the created ticket, string. E.g. resolved if you want to resolve the created ticket. The linked Investigation ticket will be resolved automatically by RTIR scripts.

  • create_investigation: if an Investigation ticket should be created (in case of RTIR workflow). true or false, boolean.

  • investigation_fields: attributes to include into investigation ticket, comma-separated string. E.g. time.source,source.ip,source.port,source.fqdn,source.url,classification.taxonomy,classification.type,classification.identifier,event_description.url,event_description.text,malware.name,protocol.application,protocol.transport.

  • description_attr: which event attribute contains text message being sent to the recipient, string. If it is not specified or not found in the event, the Investigation ticket is not going to be created. Example: extra.message.text.

REST API

Information

  • name: intelmq.bots.outputs.restapi.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: REST API is the bot responsible to send events to a REST API listener through POST

Configuration Parameters

  • auth_token: the user name / HTTP header key

  • auth_token_name: the password / HTTP header value

  • auth_type: one of: “http_basic_auth”, “http_header”

  • hierarchical_output: boolean

  • host: destination URL

  • use_json: boolean

SMTP Output Bot

Sends a MIME Multipart message containing the text and the event as CSV for every single event.

Information

  • name: intelmq.bots.outputs.smtp.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: Sends events via SMTP

Configuration Parameters

  • fieldnames: a list of field names to be included in the email, comma separated string or list of strings. If empty, no attachment is sent - this can be useful if the actual data is already in the body (parameter text) or the subject.

  • mail_from: string. Supports formatting, see below

  • mail_to: string of email addresses, comma separated. Supports formatting, see below

  • smtp_host: string

  • smtp_password: string or null, Password for authentication on your SMTP server

  • smtp_port: port

  • smtp_username: string or null, Username for authentication on your SMTP server

  • ssl: boolean

  • starttls: boolean

  • subject: string. Supports formatting, see below

  • text: string or null. Supports formatting, see below

For several strings you can use values from the string using the standard Python string format syntax. Access the event’s values with {ev[source.ip]} and similar. Any not existing fields will result in None. For example, to set the recipient(s) to the value given in the event’s source.abuse_contact field, use this as mail_to parameter: {ev[source.abuse_contact]}

Authentication is optional. If both username and password are given, these mechanism are tried: CRAM-MD5, PLAIN, and LOGIN.

Client certificates are not supported. If http_verify_cert is true, TLS certificates are checked.

SQL

Information

Configuration Parameters

The parameters marked with ‘PostgreSQL’ will be sent to libpq via psycopg2. Check the libpq parameter documentation for the versions you are using.

  • autocommit: psycopg’s autocommit mode, optional, default True

  • connect_timeout: Database connect_timeout, optional, default 5 seconds

  • engine: ‘postgresql’ or ‘sqlite’

  • database: PostgreSQL database or SQLite file

  • host: PostgreSQL host

  • jsondict_as_string: save JSONDict fields as JSON string, boolean. Default: true (like in versions before 1.1)

  • port: PostgreSQL port

  • user: PostgreSQL user

  • password: PostgreSQL password

  • sslmode: PostgreSQL sslmode, can be ‘disable’, ‘allow’, ‘prefer’ (default), ‘require’, ‘verify-ca’ or ‘verify-full’. See postgresql docs: https://www.postgresql.org/docs/current/static/libpq-connect.html#libpq-connect-sslmode

  • table: name of the database table into which events are to be inserted

PostgreSQL

You have two basic choices to run PostgreSQL:

  1. on the same machine as intelmq, then you could use Unix sockets if available on your platform

  2. on a different machine. In which case you would need to use a TCP connection and make sure you give the right connection parameters to each psql or client call.

Make sure to consult your PostgreSQL documentation about how to allow network connections and authentication in case 2.

PostgreSQL Version

Any supported version of PostgreSQL should work (v>=9.2 as of Oct 2016) [1].

If you use PostgreSQL server v >= 9.4, it gives you the possibility to use the time-zone formatting string “OF” for date-times and the GiST index for the CIDR type. This may be useful depending on how you plan to use the events that this bot writes into the database.

How to install

Use intelmq_psql_initdb to create initial SQL statements from harmonization.conf. The script will create the required table layout and save it as /tmp/initdb.sql

You need a PostgreSQL database-user to own the result database. The recommendation is to use the name intelmq. There may already be such a user for the PostgreSQL database-cluster to be used by other bots. (For example from setting up the expert/certbund_contact bot.)

Therefore if still necessary: create the database-user as postgresql superuser, which usually is done via the system user postgres:

createuser --no-superuser --no-createrole --no-createdb --encrypted --pwprompt intelmq

Create the new database:

createdb --encoding='utf-8' --owner=intelmq intelmq-events

(The encoding parameter should ensure the right encoding on platform where this is not the default.)

Now initialize it as database-user intelmq (in this example a network connection to localhost is used, so you would get to test if the user intelmq can authenticate):

psql -h localhost intelmq-events intelmq </tmp/initdb.sql

SQLite

Similarly to PostgreSQL, you can use intelmq_psql_initdb to create initial SQL statements from harmonization.conf. The script will create the required table layout and save it as /tmp/initdb.sql.

Create the new database (you can ignore all errors since SQLite doesn’t know all SQL features generated for PostgreSQL):

sqlite3 your-db.db
sqlite> .read /tmp/initdb.sql

Then, set the database parameter to the your-db.db file path.

STOMP

Information

Requirements :

Install the stomp.py library, e.g. apt install python3-stomp.py or pip install stomp.py.

You need a CA certificate, client certificate and key file from the organization / server you are connecting to. Also you will need a so called “exchange point”.

Configuration Parameters

  • exchange: The exchange to push at

  • heartbeat: default: 60000

  • message_hierarchical_output: Boolean, default: false

  • message_jsondict_as_string: Boolean, default: false

  • message_with_type: Boolean, default: false

  • port: Integer, default: 61614

  • server: Host or IP address of the STOMP server

  • single_key: Boolean or string (field name), default: false

  • ssl_ca_certificate: path to CA file

  • ssl_client_certificate: path to client cert file

  • ssl_client_certificate_key: path to client cert key file

TCP

Information

  • name: intelmq.bots.outputs.tcp.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: TCP is the bot responsible to send events to a TCP port (Splunk, another IntelMQ, etc..).

Multihreading is disabled for this bot.

Configuration Parameters

  • counterpart_is_intelmq: Boolean. If you are sending to an IntelMQ TCP collector, set this to True, otherwise e.g. with filebeat, set it to false.

  • ip: IP of destination server

  • hierarchical_output: true for a nested JSON, false for a flat JSON (when sending to a TCP collector).

  • port: port of destination server

  • separator: separator of messages, e.g. “n”, optional. When sending to a TCP collector, parameter shouldn’t be present. In that case, the output waits every message is acknowledged by “Ok” message the TCP collector bot implements.

Sending to an IntelMQ TCP collector

If you intend to link two IntelMQ instance via TCP, set the parameter counterpart_is_intelmq to true. The bot then awaits an “Ok” message to be received after each message is sent. The TCP collector just sends “Ok” after every message it gets.

Templated SMTP

Sends a MIME Multipart message built from an event and static text using Jinja2 templates.

Information

  • name: intelmq.bots.outputs.templated_smtp.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: Sends events via SMTP

Requirements

Install the required jinja2 library:

pip3 install -r intelmq/bots/collectors/templated_smtp/REQUIREMENTS.txt

Configuration Parameters

Parameters:

  • attachments: list of objects with structure:

    - content-type: string, templated, content-type to use.
      text: string, templated, attachment text.
      name: string, templated, filename of attachment.
    
  • body: string, optional, templated, body text. The default body template prints every field in the event except ‘raw’, in undefined order, one field per line, as “field: value”.

  • mail_from: string, templated, sender address.

  • mail_to: string, templated, recipient addresses, comma-separated.

  • smtp_host: string, optional, default “localhost”, hostname of SMTP server.

  • smtp_password: string, default null, password (if any) for authenticated SMTP.

  • smtp_port: integer, default 25, TCP port to connect to.

  • smtp_username: string, default null, username (if any) for authenticated SMTP.

  • tls: boolean, default false, whether to use use SMTPS. If true, also set smtp_port to the SMTPS port.

  • starttls: boolean, default true, whether to use opportunistic STARTTLS over SMTP.

  • subject: string, optional, default “IntelMQ event”, templated, e-mail subject line.

  • verify_cert: boolean, default true, whether to verify the server certificate in STARTTLS or SMTPS.

Authentication is attempted only if both username and password are specified.

Templates are in Jinja2 format with the event provided in the variable “event”. E.g.:

mail_to: "{{ event['source.abuse_contact'] }}"

See the Jinja2 documentation at https://jinja.palletsprojects.com/ .

Attachments are template strings, especially useful for sending structured data. E.g. to send a JSON document including “malware.name” and all other fields starting with “source.”:

attachments:
  - content-type: application/json
    text: |
      {
        "malware": "{{ event['malware.name'] }}",
        {%- set comma = joiner(", ") %}
        {%- for key in event %}
           {%- if key.startswith('source.') %}
        {{ comma() }}"{{ key }}": "{{ event[key] }}"
           {%- endif %}
        {%- endfor %}
      }
    name: report.json

You are responsible for making sure that the text produced by the template is valid according to the content-type.

If you are migrating from the SMTP output bot that produced CSV format attachments, use the following configuration to produce a matching format:

attachments:
  - content-type: text/csv
    text: |
      {%- set fields = ["classification.taxonomy", "classification.type", "classification.identifier", "source.ip", "source.asn", "source.port"] %}
      {%- set sep = joiner(";") %}
      {%- for field in fields %}{{ sep() }}{{ field }}{%- endfor %}
      {% set sep = joiner(";") %}
      {%- for field in fields %}{{ sep() }}{{ event[field] }}{%- endfor %}
    name: event.csv

Touch

Information

  • name: intelmq.bots.outputs.touch.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: Touches a file for every event received.

Configuration Parameters

  • path: Path to the file to touch.

UDP

Information

  • name: intelmq.bots.outputs.udp.output

  • lookup: no

  • public: yes

  • cache (redis db): none

  • description: Output Bot that sends events to a remote UDP server.

Multihreading is disabled for this bot.

Configuration Parameters

  • field_delimiter: If the format is ‘delimited’ this will be added between fields. String, default: “|”

  • format: Can be ‘json’ or ‘delimited’. The JSON format outputs the event ‘as-is’. Delimited will deconstruct the event and print each field:value separated by the field delimit. See examples below.

  • header: Header text to be sent in the UDP datagram, string.

  • keep_raw_field: boolean, default: false

  • udp_host: Destination’s server’s Host name or IP address

  • udp_port: Destination port

Examples of usage

Consider the following event:

{"raw": "MjAxNi8wNC8yNV8xMTozOSxzY2hpenppbm8ub21hcmF0aG9uLmNvbS9na0NDSnVUSE0vRFBlQ1pFay9XdFZOSERLbC1tWFllRk5Iai8sODUuMjUuMTYwLjExNCxzdGF0aWMtaXAtODUtMjUtMTYwLTExNC5pbmFkZHIuaXAtcG9vbC5jb20uLEFuZ2xlciBFSywtLDg5NzI=", "source": {"asn": 8972, "ip": "85.25.160.114", "url": "http://schizzino.omarathon.com/gkCCJuTHM/DPeCZEk/WtVNHDKl-mXYeFNHj/", "reverse_dns": "static-ip-85-25-160-114.inaddr.ip-pool.com"}, "classification": {"type": "malware-distribution"}, "event_description": {"text": "Angler EK"}, "feed": {"url": "http://www.malwaredomainlist.com/updatescsv.php", "name": "Malware Domain List", "accuracy": 100.0}, "time": {"observation": "2016-04-29T10:59:34+00:00", "source": "2016-04-25T11:39:00+00:00"}}

With the following Parameters:

  • field_delimiter : |

  • format : json

  • Header : header example

  • keep_raw_field : true

  • ip : 127.0.0.1

  • port : 514

Resulting line in syslog:

Apr 29 11:01:29 header example {"raw": "MjAxNi8wNC8yNV8xMTozOSxzY2hpenppbm8ub21hcmF0aG9uLmNvbS9na0NDSnVUSE0vRFBlQ1pFay9XdFZOSERLbC1tWFllRk5Iai8sODUuMjUuMTYwLjExNCxzdGF0aWMtaXAtODUtMjUtMTYwLTExNC5pbmFkZHIuaXAtcG9vbC5jb20uLEFuZ2xlciBFSywtLDg5NzI=", "source": {"asn": 8972, "ip": "85.25.160.114", "url": "http://schizzino.omarathon.com/gkCCJuTHM/DPeCZEk/WtVNHDKl-mXYeFNHj/", "reverse_dns": "static-ip-85-25-160-114.inaddr.ip-pool.com"}, "classification": {"type": "malware-distribution"}, "event_description": {"text": "Angler EK"}, "feed": {"url": "http://www.malwaredomainlist.com/updatescsv.php", "name": "Malware Domain List", "accuracy": 100.0}, "time": {"observation": "2016-04-29T10:59:34+00:00", "source": "2016-04-25T11:39:00+00:00"}}

With the following Parameters:

  • field_delimiter : |

  • format : delimited

  • Header : IntelMQ-event

  • keep_raw_field : false

  • ip : 127.0.0.1

  • port : 514

Resulting line in syslog:

Apr 29 11:17:47 localhost IntelMQ-event|source.ip: 85.25.160.114|time.source:2016-04-25T11:39:00+00:00|feed.url:http://www.malwaredomainlist.com/updatescsv.php|time.observation:2016-04-29T11:17:44+00:00|source.reverse_dns:static-ip-85-25-160-114.inaddr.ip-pool.com|feed.name:Malware Domain List|event_description.text:Angler EK|source.url:http://schizzino.omarathon.com/gkCCJuTHM/DPeCZEk/WtVNHDKl-mXYeFNHj/|source.asn:8972|classification.type:malware-distribution|feed.accuracy:100.0