Shipyard ships a Makefile.inc file which defines these basic targets:
If your project uses Shipyard then it has all these targets and supports all the variables these targets support.
Any variables supported by these targets can be assigned on the make
command line.
Many targets support variables that influence how each target behaves.
SETTINGS
: Settings file that specifies a topology for deployment.PROVIDER
: Cloud provider for the infrastructure (defaults to kind
).GOLBALNET
: When true, deploys the clusters with overlapping IPs (defaults to false
).DEBUG_PRINT
: When true, outputs debug information for Shipyard’s scripts (defaults to true
).Creates a kind-based multi-cluster environment with just the default Kubernetes deployment:
make clusters
K8S_VERSION
: Determines the Kubernetes version that gets deployed (defaults to 1.24
).Deploys Submariner components in a kind-based cluster environment (if one isn’t created yet, this target will first invoke the clusters
target to do so):
make deploy
CABLE_DRIVER
: The cable driver used by Submariner (defaults to libreswan
).DEPLOYTOOL
: The tool used to deploy Submariner itself (defaults to operator
).LIGHTHOUSE
: Deploys Lighthouse in addition to the basic Submariner deployment (defaults to false
).Runs end to end testing on the deployed environment (if one isn’t created yet, this target will first invoke the deploy
target to do so).
The tests are taken from the project, unless it has no specific end to end tests, in which case generic testing using subctl verify
is
run.
make e2e
To clean up all the kind clusters deployed in any of the previous steps, use:
make clean-clusters
This command will remove the clusters and any resources that might’ve been left in docker that are not needed any more (images, volumes, etc).
To clean up all generated files, use:
make clean-generated
This will remove any file which can be re-generated and doesn’t need to be tracked.
To clean everything up, use:
make clean
This removes any running clusters and all generated files.