falcodump1
Dump log data to a file using a falco source plugin.
Usage
Common options
falcodump [--help]
[--version]
[--plugin-api-version]
[--extcap-interfaces]
[--extcap-dlts]
[--extcap-interface=<interface>]
[--extcap-config]
[--extcap-capture-filter=<capture
filter>]
[--capture]
[--fifo=<path
to
file
or
pipe>]
[--plugin-source=<source
path
or
URL>]
[--log-level=<log
level>]
[--log-file=<path
to
file>]
System call
options
[--include-capture-processes=<TRUE
or
FALSE>]
[--include-switch-calls=<TRUE
or
FALSE>]
CloudTrail plugin
options
[--cloudtrail-s3downloadconcurrency=<number
of
concurrent
downloads>]
[--cloudtrail-s3interval=<timeframe>]
[--cloudtrail-s3accountlist=<comma
separated
account
IDs>]
[--cloudtrail-sqsdelete=<true
or
false>]
[--cloudtrail-useasync=<true
or
false>]
[--cloudtrail-uses3sns=<true
or
false>]
[--cloudtrail-aws-region=<AWS
region>]
[--cloudtrail-aws-profile=<AWS
profile>]
[--cloudtrail-aws-config=<path>]
[--cloudtrail-aws-credentials=<path
to
file>]
Overview
falcodump is an extcap tool that allows one to capture log messages from cloud providers.
Each plugin is listed as a separate interface. For example, the AWS CloudTrail plugin is listed as “cloudtrail”.
Options
-
--help -
Print program arguments. This will also list the configuration arguments for each plugin.
-
--version -
Print the program version.
-
--plugin-api-version -
Print the Falco plugin API version.
-
--extcap-interfaces -
List the available interfaces.
-
--extcap-interface=<interface> -
Use the specified interface.
-
--extcap-dlts -
List the DLTs of the specified interface.
-
--extcap-config -
List the configuration options of specified interface.
--extcap-capture-filter=<capture filter>
The capture filter. Must be a valid Sysdig / Falco filter.
-
--capture -
Start capturing from the source specified by --plugin-source via the specified interface and write raw packet data to the location specified by --fifo.
--fifo=<path to file or pipe>
Save captured packet to file or send it through pipe.
--plugin-source=<source path or URL>
Capture from the specified location.
-
--log-level -
Set the log level
-
--log-file -
Set a log file to log messages in addition to the console
Options
-
--include-capture-processes -
Include system calls for capture processes (falcodump, dumpcap, and Logray) if TRUE. Defaults to FALSE.
-
--include-switch-calls -
Include "switch" calls if TRUE. Defaults to FALSE.
Plugins
cloudtrail (AWS CloudTrail)
--cloudtrail-s3downloadconcurrency
Controls the number of background goroutines used to download S3 files (Default: 32)
--cloudtrail-s3interval
Download log files over the specified interval (Default: no interval)
--cloudtrail-s3accountlist
If source is an organization CloudTrail S3 bucket download log files for all specified account IDs (Default: no account IDs)
--cloudtrail-sqsdelete
If true then the plugin will delete SQS messages from the queue immediately after receiving them (Default: true)
--cloudtrail-useasync
If true then async extraction optimization is enabled (Default: true)
--cloudtrail-uses3sns
If true then the plugin will expect SNS messages to originate from S3 instead of directly from Cloudtrail (Default: false)
--cloudtrail-aws-profile
If non-empty overrides the AWS shared configuration profile (e.g. 'default') and environment variables such as AWS_PROFILE (Default: empty)
--cloudtrail-aws-region
If non-empty overrides the AWS region specified in the profile (e.g. 'us-east-1') and environment variables such as AWS_REGION (Default: empty)
--cloudtrail-aws-config
If non-empty overrides the AWS shared configuration filepath (e.g. ~/.aws/config) and env variables such as AWS_CONFIG_FILE (Default: empty)
--cloudtrail-aws-credentials
If non-empty overrides the AWS shared credentials filepath (e.g. ~/.aws/credentials) and env variables such as AWS_SHARED_CREDENTIALS_FILE (Default: empty)
CloudTrail sources can be S3 buckets or SQS queue URLs. S3 bucket URLs have the form
's3://bucket_name/prefix/AWSLogs/account-id/CloudTrail/region/year/month/day'
For organization CloudTrail the S3 bucket URL can be
's3://bucket_name/prefix/AWSLogs/org-id/account-id/CloudTrail/region/year/month/day'
The region, year, month, and day components can be omitted in order to fetch more or less data. For example, the source 's3://mybucket/AWSLogs/012345678/CloudTrail/us-west-2/2023' will fetch all CloudWatch logs for the year 2023.
If the URL ends with 'account-id/' or 'account-id/CloudTrail/' (for example 's3://mybucket/AWSLOGS/012345678912/') the option '--cloudtrail-s3interval' can be used to define the time frame. A s3interval of '1d' for example would get all events of the last 24 hours from all available regions. A s3interval of '2w-1w' would get all events from all regions from two weeks ago up to one week ago. The s3invterval can also be defined as a RFC 3339-style timestamp like '2024-02-29T18:07:17Z' or '2024-02-29T00:00:00Z-2024-03-01T23:59:59Z'.
If the URL ends with 'AWSLogs/org-id' option '--cloudtrail-s3accountlist' can be used to specify account IDs. This can be combined with '--cloudtrail-s3interval'. A source like 's3://my-org-bucket/AWSLogs/o-123abc/' with '--cloudstrail-s3accountlist' set to '123456789012,987654321098' and '--cloudtrail-s3interval' set to '30m' would get all events of the last 30min from all regions for accounts 123456789012 and 987654321098.
If source URL is the organization CloudTrail bucket (like 's3://my-org-bucket/AWSLogs/o-123abc') and '--s3accountlist' is not set the plugin iterates over all accounts (limited by '--s3interval' if set). Attention: Depending on the size of the organization and the time interval, this can take a long time.
The cloudtrail plugin uses the AWS SDK for Go, which can obtain profile, region, and credential settings from a set of standard environment variables and configuration files <https://aws.github.io/aws-sdk-go-v2/docs/configuring-sdk/>. Falcodump will show a list of locally configured profiles and the current regions, and will let you supply a custom value as well.
More information is available in the README <https://github.com/falcosecurity/plugins/blob/master/plugins/cloudtrail/README.md> of the CloudTrail plugin.
Examples
To see program arguments:
falcodump --help
To see program version:
falcodump --version
To see interfaces:
falcodump --extcap-interfaces
Only one interface (falcodump) is supported.
Example output
interface {value=cloudtrail}{display=Falco plugin}
To see interface DLTs:
falcodump --extcap-interface=cloudtrail --extcap-dlts
Example output
dlt {number=147}{name=cloudtrail}{display=USER0}
To see interface configuration options:
falcodump --extcap-interface=cloudtrail --extcap-config
Example output
arg {number=0}{call=--plugin-source}{display=Plugin source}{type=string}{tooltip=The plugin data source. This us usually a URL.}{placeholder=Enter a source URL…}{required=true}{group=Capture}
arg {number=1}{call=cloudtrail-s3downloadconcurrency}{display=s3DownloadConcurrency}{type=integer}{default=1}{tooltip=Controls the number of background goroutines used to download S3 files (Default: 1)}{group=Capture}
arg {number=2}{call=cloudtrail-sqsdelete}{display=sqsDelete}{type=boolean}{default=true}{tooltip=If true then the plugin will delete sqs messages from the queue immediately after receiving them (Default: true)}{group=Capture}
arg {number=3}{call=cloudtrail-useasync}{display=useAsync}{type=boolean}{default=true}{tooltip=If true then async extraction optimization is enabled (Default: true)}{group=Capture}
To capture AWS CloudTrail events from an S3 bucket:
falcodump --extcap-interface=cloudtrail --fifo=/tmp/cloudtrail.pcap --plugin-source=s3://aws-cloudtrail-logs.../CloudTrail/us-east-2/... --capture
or:
falcodump --capture --extcap-interface cloudtrail --fifo ~/cloudtrail.pcap --plugin-source s3://my-cloudtrail-bucket/AWSLogs/o-abc12345/123456789012/ --cloudtrail-s3downloadconcurrency 32 --cloudtrail-s3interval 5d-2d --cloudtrail-aws-region eu-west-1
Note
CTRL + C should be used to stop the capture in order to ensure clean termination.
Notes
falcodump is part of the Logray distribution. The latest version of Logray can be found at <https://www.wireshark.org>.
HTML versions of the Wireshark project man pages are available at <https://www.wireshark.org/docs/man-pages>.