You can decide at build time which components of the Agent you want to find in
the final artifact. By default, all the components are picked up, so if you want
to replicate the same configuration of the Agent distributed via system packages,
all you have to do is deva agent.build
.
To pick only certain components you have to invoke the task like this:
deva agent.build --build-include=zstd,etcd,python
Conversely, if you want to exclude something:
deva agent.build --build-exclude=systemd,python
This is the complete list of the available components:
apm
: make the APM agent execution available. For information on building the trace agent, see the trace agent README.consul
: enable consul as a configuration storepython
: embed the Python interpreter.docker
: add Docker support (required by AutoDiscovery).ec2
: enable EC2 hostname detection and metadata collection.etcd
: enable Etcd as a configuration store.gce
: enable GCE hostname detection and metadata collection.jmx
: enable the JMX-fetch bridge.kubelet
: enable kubelet tag collectionlog
: enable the log agentprocess
: enable the process agentzk
: enable Zookeeper as a configuration store.zstd
: use Zstandard instead of Zlib.systemd
: enable systemd journal log collectionnetcgo
: force the use of the CGO resolver. This will also have the effect of making the binary non-staticsecrets
: enable secrets support in configuration files (see documentation here)clusterchecks
: enable cluster-level checkscri
: add support for the CRI integrationcontainerd
: add support for the containerd integrationkubeapiserver
: enable interaction with Kubernetes API server (required by the cluster Agent)
Please note you might need to provide some extra dependencies in your dev environment to build certain bits (see development environment).
Also note that the trace agent needs to be built and run separately. For more information, see the trace agent README.
We use pkg-config
to make compilers and linkers aware of Python. The required .pc files are
provided automatically when building python through omnibus.
Building an Agent Docker image from scratch through an embedded build is a slow process. You can quickly test a change or bug fix in a containerized environment (such as Docker, Kubernetes, or ECS).
One way to do this is to patch the Agent binary from an official Docker image, with a Dockerfile:
FROM datadog/agent:<AGENT_VERSION>
COPY agent /opt/datadog-agent/bin/agent/agent
For this to work properly, two things are important:
- Your change needs to be done on top of the
<AGENT_VERSION>
tag from the DataDog repository. - You need to run the invoke task with the proper embedded path
deva -e agent.build -e /opt/datadog-agent/embedded
.
Note: This makes invoke
install the build's artifacts in the /opt/datadog-agent/embedded
folder. Make sure the folder exists and the current user has write permissions.