A simple daemon for connecting Falco
to your ecosystem. It takes a Falco
events and
forward them to different outputs in a fan-out way.
It works as a single endpoint for as many as you want Falco
instances :
Falcosidekick
manages a large variety of outputs with different purposes.
[!NOTE] Follow the links to get the configuration of each output.
falcosidekick
)falcosidekick
)falcosidekick
)falcosidekick
)Run the daemon as any other daemon in your architecture (systemd, k8s deployment, swarm service, ...).
Use the environment variables to set up the outputs:
docker run -d -p 2801:2801 -e SLACK_WEBHOOKURL=XXXX -e DATADOG_APIKEY=XXXX falcosecurity/falcosidekick
Download the latest release:
VER=$(curl --silent -qI https://github.com/falcosecurity/falcosidekick/releases/latest | awk -F '/' '/^location/ {print substr($NF, 1, length($NF)-1)}')
wget -c https://github.com/falcosecurity/falcosidekick/releases/download/${VER}/falcosidekick_${VER}_linux_arm64.tar.gz -O - | tar -xz
or
wget -c https://github.com/falcosecurity/falcosidekick/releases/download/${VER}/falcosidekick_${VER}_linux_amd64.tar.gz -O - | tar -xz
chmod +x falcosidekick
sudo mv falcosidekick /usr/local/bin/
Create the /etc/falcosidekick/config.yaml
file, see Configuration.
Create the systemd unit files /etc/systemd/system/falcosidekick.service
:
sudo touch /etc/systemd/system/falcosidekick.service
sudo chmod 664 /etc/systemd/system/falcosidekick.service
[Unit]
Description=Falcosidekick
After=network.target
StartLimitIntervalSec=0
[Service]
Type=simple
Restart=always
RestartSec=1
ExecStart=/usr/local/bin/falcosidekick -c /etc/falcosidekick/config.yaml
[Install]
WantedBy=default.target
Reload systemd
and start Falcosidekick
:
sudo systemctl daemon-reload
sudo systemctl enable falcosidekick
sudo systemctl start falcosidekick
Check if Falcosidekick
runs:
curl localhost:2801/healthz
See https://github.com/falcosecurity/charts/blob/master/charts/falcosidekick/README.md
helm repo add falcosecurity https://falcosecurity.github.io/charts
helm repo update
helm install falcosidekick --set config.debug=true falcosecurity/falcosidekick
[!NOTE] You can also deploy
falcosidekick
as a dependency of thefalco
chart, the settings for the communication between falco andfalcosidekick
are automatically set. Just prefix allfalcosidekick
settings withfalcosidekick.
:helm repo add falcosecurity https://falcosecurity.github.io/charts helm repo update
helm install falco --set falcosidekick.enabled=true falcosecurity/falco
### Connect Falco
To connect Falco with Falcosidekick, you need to change it configuration as following:
#### with falco.yaml
If managing _falco.yaml_ manually, set this:
```yaml
json_output: true
json_include_output_property: true
http_output:
enabled: true
url: "http://localhost:2801/"
If installing falco
with Helm
, set this (adapted to your environment) in
your values.yaml :
falcosidekick:
enabled: true
or
jsonOutput: true
jsonIncludeOutputProperty: true
httpOutput:
enabled: true
url: "http://falcosidekick:2801/"
or
jsonOutput: true
jsonIncludeOutputProperty: true
programOutput:
enabled: true
keepAlive: false
program: "curl -d @- falcosidekick:2801/"
Configuration is made by file (yaml) and env vars, both can be used but env vars override values from file.
See config_example.yaml :
#listenaddress: "" # ip address to bind falcosidekick to (default: "" meaning all addresses)
#listenport: 2801 # port to listen for daemon (default: 2801)
debug: false # if true all outputs will print in stdout the payload they send (default: false)
customfields: # custom fields are added to falco events, if the value starts with % the relative env var is used
# Akey: "AValue"
# Bkey: "BValue"
# Ckey: "CValue"
templatedfields: # templated fields are added to falco events and metrics, it uses Go template + output_fields values
# Dkey: '{{ or (index . "k8s.ns.labels.foo") "bar" }}'
customtags: # custom tags are added to the falco events, if the value starts with % the relative env var is used
# - tagA
# - tagB
# bracketreplacer: "_" # if not empty, replace the brackets in keys of Output Fields
outputFieldFormat: "<timestamp>: <priority> <output> <custom_fields> <templated_fields>" # if not empty, allow to change the format of the output field. (default: "<timestamp>: <priority> <output>")
mutualtlsfilespath: "/etc/certs" # folder which will used to store client.crt, client.key and ca.crt files for mutual tls for outputs, will be deprecated in the future (default: "/etc/certs")
mutualtlsclient: # takes priority over mutualtlsfilespath if not emtpy
certfile: "/etc/certs/client/client.crt" # client certification file
keyfile: "/etc/certs/client/client.key" # client key
cacertfile: "/etc/certs/client/ca.crt" # for server certification
tlsclient:
cacertfile: "/etc/certs/client/ca.crt" # CA certificate file for server certification on TLS connections, appended to the system CA pool if not empty
tlsserver:
deploy: false # if true, TLS server will be deployed instead of HTTP
certfile: "/etc/certs/server/server.crt" # server certification file
keyfile: "/etc/certs/server/server.key" # server key
mutualtls: false # if true, mTLS server will be deployed instead of TLS, deploy also has to be true
cacertfile: "/etc/certs/server/ca.crt" # for client certification if mutualtls is true
notlsport: 2810 # port to serve http server serving selected endpoints (default: 2810)
notlspaths: # if not empty, and tlsserver.deploy is true, a separate http server will be deployed for the specified endpoints
- "/ping"
# - "/metrics"
# - "/healthz"
[!NOTE] For the confiuration of the outputs, see the docs.
Usage :
usage: falcosidekick [<flags>]
Flags:
--help Show context-sensitive help (also try --help-long and --help-man).
-c, --config-file=CONFIG-FILE config file
Different endpoints (handlers) are available :
/
: main and default handler, your falco config must be configured to use it/ping
: you will get a pong
as answer, useful to test if falcosidekick is running and its port is opened (for healthcheck purpose for example). This endpoint is deprecated and it will be removed in 3.0.0
./healthz
: you will get a HTTP status code 200
response as answer, useful to test if falcosidekick is running and its port is opened (for healthcheck or purpose for example)/test
: (for debug only) send a test event to all enabled outputs./debug/vars
: get statistics from daemon (in JSON format), it uses classic expvar
package and some custom values are added/metrics
: prometheus endpoint, for scraping metrics about events and falcosidekick
All logs are sent to stdout
.
2019/05/10 14:32:06 [INFO] : Enabled Outputs : Slack Datadog
Outputs with mutualtls
enabled in their configuration require the client.crt, client.key and ca.crt filepaths to be configured in the mutualtlsclient_certfile, mutualtlsclient_keyfile and mutualtlsclient_cacertfile global parameter.
docker run -d -p 2801:2801 -e MUTUALTLSCLIENT_CERTFILE=/etc/certs/client/client.crt -e MUTUALTLSCLIENT_KEYFILE=/etc/certs/client/client.key -e MUTUALTLSCLIENT_CACERTFILE=/etc/certs/client/ca.crt -e ALERTMANAGER_HOSTPORT=https://XXXX -e ALERTMANAGER_MUTUALTLS=true -e INFLUXDB_HOSTPORT=https://XXXX -e INFLUXDB_MUTUALTLS=true -e WEBHOOK_ADDRESS=XXXX -v /localpath/myclientcert.crt:/etc/certs/client/client.crt -v /localpath/myclientkey.key:/etc/certs/client/client.key -v /localpath/ca.crt:/etc/certs/client/ca.crt falcosecurity/falcosidekick
Alternately the path where the client.crt, client.key and ca.crt files are stored can be configured in mutualtlsfilespath global parameter. (Important: file names must be preserved)
docker run -d -p 2801:2801 -e MUTUALTLSFILESPATH=/etc/certs -e ALERTMANAGER_HOSTPORT=https://XXXX -e ALERTMANAGER_MUTUALTLS=true -e INFLUXDB_HOSTPORT=https://XXXX -e INFLUXDB_MUTUALTLS=true -e WEBHOOK_ADDRESS=XXXX -v /localpath/myclientcert.crt:/etc/certs/client.crt -v /localpath/myclientkey.key:/etc/certs/client.key -v /localpath/ca.crt:/etc/certs/ca.crt falcosecurity/falcosidekick
In above example, the same client certificate will be used for both Alertmanager & InfluxDB outputs which have mutualtls flag set to true.
The daemon exposes the common Golang metrics and some custom values in JSON format. It's useful for monitoring purpose.
The daemon exposes a prometheus
endpoint on URI /metrics
.
See the docs for more info.
The daemon is able to push its metrics to a StatsD/DogstatsD server. See Configuration section for how-to.
See the statsd docs and dogstastd docs for more info.
Run you daemon and try (from Falco's documentation):
curl -XPOST "http://localhost:2801/" -d'{"output":"16:31:56.746609046: Error File below a known binary directory opened for writing (user=root command=touch /bin/hack file=/bin/hack)","hostname": "localhost", "priority":"Error","rule":"Write below binary dir","time":"2019-05-17T15:31:56.746609046Z", "output_fields": {"evt.time":1507591916746609046,"fd.name":"/bin/hack","proc.cmdline":"touch /bin/hack","user.name":"root"}}'
make falcosidekick
Create a debug event
curl -X POST -H "Content-Type: application/json" -H "Accept: application/json" localhost:2801/test
make test
With Coverage
make test-coverage
Thomas Labarussias (https://github.com/Issif)