Configure calicoctl

Big picture

Learn how to configure the calicoctl CLI tool for your cluster.

Value

The calicoctl CLI tool provides helpful administrative commands for interacting with a Calico cluster.

Concepts

calicoctl vs kubectl

In previous releases, calicoctl has been required to manage Calico API resources in the projectcalico.org/v3 API group. The calicoctl CLI tool provides important validation and defaulting on these APIs.

In newer releases, the Calico API server performs that defaulting and validation server-side, exposing the same API semantics without a dependency on calicoctl. For this reason, we recommend installing the Calico API server and using kubectl instead of calicoctl for most operations.

calicoctl is still required for the following subcommands:

calicoctl is also required for non-Kubernetes platforms such as OpenStack.

Default calicoctl behavior

Most calicoctl commands require access to the Calico datastore. By default, calicoctl will attempt to read from the Kubernetes API based on the default kubeconfig.

How to

Configure access using a Configuration file

By default, calicoctl will look for a configuration file at /etc/calico/calicoctl.cfg. You can override this using the --config option with commands that require datastore access. The file can be in either YAML or JSON format. It must be valid and readable by calicoctl. For example:

  1. apiVersion: projectcalico.org/v3
  2. kind: CalicoAPIConfig
  3. metadata:
  4. spec:
  5. datastoreType: "etcdv3"
  6. etcdEndpoints: "http://etcd1:2379,http://etcd2:2379"
  7. ...

Configure access using environment variables

If calicoctl cannot locate, read, or access a configuration file, it will check a specific set of environment variables.

Refer to the section that corresponds to your datastore type for a full set of options and examples.

Configure calicoctl - 图1note

When running calicoctl inside a container, any environment variables and configuration files must be passed to the container so they are available to the process inside. It can be useful to keep a running container (that sleeps) configured for your datastore, then it is possible to exec into the container and have an already configured environment.