Skip to main content
Version: v4.21

velociraptor

Submits VQL to a Velociraptor server and returns the response as events.

Synopsis

velociraptor [-n|--request-name <string>] [-o|--org-id <string>]
[-r|--max-rows <uint64>] [-s|--subscribe <artifact>]
[-q|--query <vql>] [-w|--max-wait <duration>]
[--profile <profile>]

Description

The velociraptor source operator provides a request-response interface to a Velociraptor server:

The pipeline operator is the client and it establishes a connection to a Velociraptor server. The client request contains a query written in the Velociraptor Query Language (VQL), a SQL-inspired language with a SELECT .. FROM .. WHERE structure.

You can either send a raw VQL query via velociraptor --query "<vql>" to a server and processs the response, or hook into a continuous feed of artifacts via velociraptor --subscribe <artifact>. Whenever a hunt runs that contains this artifact, the server will forward it to the pipeline and emit the artifact payload in the response field HuntResults.

All Velociraptor client-to-server communication is mutually authenticated and encrypted via TLS certificates. This means you must provide client-side certificate, which you can generate as follows. (Velociraptor ships as a static binary that we refer to as velociraptor-binary here.)

  1. Create a server configuration server.yaml:

    velociraptor-binary config generate > server.yaml
  2. Create an API client:

    velociraptor-binary -c server.yaml config api_client --name tenzir client.yaml

    Copy the generated client.yaml to your Tenzir plugin configuration directory as velociraptor.yaml so that the operator can find it:

    cp client.yaml /etc/tenzir/plugin/velociraptor.yaml
  3. Run the frontend with the server configuration:

    velociraptor-binary -c server.yaml frontend

Now you are ready to run VQL queries!

-n|--request-name <string>

An identifier for the request to the Velociraptor server.

Defaults to a randoum UUID.

-o|--org-id <string>

The ID of the Velociraptor organization.

Defaults to root.

-q|--query <vql>

The VQL query string.

-r|--max-rows <uint64>

The maxium number of rows to return in a the stream gRPC messages returned by the server.

Defaults to 1,000.

-s|--subscribe <artifact>

Subscribes to a flow artifact.

This option generates a larger VQL expression under the hood that creates one event per flow and artifact. The response contains a field HuntResult that contains the result of the hunt.

-w|--max-wait <duration>

Controls how long to wait before releasing a partial result set.

Defaults to 1 sec.

--profile <profile>

Specifies the configuration profile for the Velociraptor instance. This enables connecting to multiple Velociraptor instances from the same Tenzir node.

To use profiles, edit your velociraptor.yaml configuration like this, where <config> refers to the contents of the configuration file created by Velociraptor, and <profile> to the desired profile name.

# before
<config>
 
# after
profiles:
  <profile>:
    <config>

If profiles are defined, the operator defaults to the first profile.

Examples

Show all processes:

velociraptor --query "select * from pslist()"

Subscribe to a hunt flow that contains the Windows artifact:

velociraptor --subscribe Windows