runner

package
v0.4.0 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Apr 9, 2020 License: Apache-2.0, MIT Imports: 50 Imported by: 0

Documentation

Index

Constants

View Source 
const (

	// magic values that we monitor on the Testground runner side to detect when Testground
	// testplan instances are initialised and at the stage of actually running a test
	// check sdk/sync for more information
	NetworkInitialisationSuccessful = "network initialisation successful"
	NetworkInitialisationFailed     = "network initialisation failed"
)
View Source 
const (
	Error eventType = iota
	Start
	Ok
	Fail
	Crash
	Incomplete
	Message
	Metric
	Other
	InternalErr
)
View Source 
const InfraMaxFilesUlimit int64 = 1048576

Variables

This section is empty.

Functions

This section is empty.

Types

type ClusterK8sRunner 

type ClusterK8sRunner struct {
	// contains filtered or unexported fields
}

ClusterK8sRunner is a runner that creates a Docker service to launch as many replicated instances of a container as the run job indicates.

func (*ClusterK8sRunner) CollectOutputs 

func (c *ClusterK8sRunner) CollectOutputs(ctx context.Context, input *api.CollectionInput, ow *rpc.OutputWriter) error

func (*ClusterK8sRunner) CompatibleBuilders 

func (*ClusterK8sRunner) CompatibleBuilders() []string

func (*ClusterK8sRunner) ConfigType 

func (*ClusterK8sRunner) ConfigType() reflect.Type

func (*ClusterK8sRunner) Healthcheck added in v0.3.0 

func (c *ClusterK8sRunner) Healthcheck(_ context.Context, engine api.Engine, ow *rpc.OutputWriter, fix bool) (*api.HealthcheckReport, error)

func (*ClusterK8sRunner) ID 

func (*ClusterK8sRunner) ID() string

func (*ClusterK8sRunner) Run 

func (c *ClusterK8sRunner) Run(ctx context.Context, input *api.RunInput, ow *rpc.OutputWriter) (*api.RunOutput, error)

func (*ClusterK8sRunner) TerminateAll added in v0.2.0 

func (c *ClusterK8sRunner) TerminateAll(_ context.Context, ow *rpc.OutputWriter) error

Terminates all pods for with the label testground.purpose: plan This command will remove all plan pods in the cluster.

type ClusterK8sRunnerConfig 

type ClusterK8sRunnerConfig struct {
	// LogLevel sets the log level in the test containers (default: not set).
	LogLevel string `toml:"log_level"`

	KeepService bool `toml:"keep_service"`

	// Resources requested for each pod from the Kubernetes cluster
	PodResourceMemory string `toml:"pod_resource_memory"`
	PodResourceCPU    string `toml:"pod_resource_cpu"`
}

ClusterK8sRunnerConfig is the configuration object of this runner. Boolean values are expressed in a way that zero value (false) is the default setting.

type ClusterSwarmRunner 

type ClusterSwarmRunner struct{}

ClusterSwarmRunner is a runner that creates a Docker service to launch as many replicated instances of a container as the run job indicates.

func (*ClusterSwarmRunner) CollectOutputs 

func (*ClusterSwarmRunner) CollectOutputs(ctx context.Context, input *api.CollectionInput, ow *rpc.OutputWriter) error

func (*ClusterSwarmRunner) CompatibleBuilders 

func (*ClusterSwarmRunner) CompatibleBuilders() []string

func (*ClusterSwarmRunner) ConfigType 

func (*ClusterSwarmRunner) ConfigType() reflect.Type

func (*ClusterSwarmRunner) ID 

func (*ClusterSwarmRunner) ID() string

func (*ClusterSwarmRunner) Run 

func (*ClusterSwarmRunner) Run(ctx context.Context, input *api.RunInput, ow *rpc.OutputWriter) (*api.RunOutput, error)

TODO runner option to keep containers alive instead of deleting them after the test has run.

type ClusterSwarmRunnerConfig 

type ClusterSwarmRunnerConfig struct {
	// LogLevel sets the log level in the test containers (default: not set).
	LogLevel string `toml:"log_level"`

	// Background avoids tailing the output of containers, and displaying it as
	// log messages (default: true).
	Background bool `toml:"background"`

	// DockerEndpoint is the URL of the docker swarm manager endpoint, e.g.
	// "tcp://manager:2376"
	DockerEndpoint string `toml:"docker_endpoint"`

	// DockerTLS indicates whether client TLS is enabled.
	DockerTLS bool `toml:"docker_tls"`

	// DockerTLSCACertPath is the path to the CA Certificate. Only used if
	// DockerTLS = true.
	DockerTLSCACertPath string `toml:"docker_tls_ca_cert_path"`

	// DockerTLSCertPath is the path to our client cert, signed by the CA. Only
	// used if DockerTLS = true.
	DockerTLSCertPath string `toml:"docker_tls_cert_path"`

	// DockerTLSKeyPath is our private key. Only used if DockerTLS = true.
	DockerTLSKeyPath string `toml:"docker_tls_key_path"`

	// KeepService keeps the service after all instances have finished and
	// all logs have been piped. Only used when running in foreground mode
	// (default is background mode).
	KeepService bool `toml:"keep_service"`
}

ClusterSwarmRunnerConfig is the configuration object of this runner. Boolean values are expressed in a way that zero value (false) is the default setting.

type FakeWriterAt 

type FakeWriterAt struct {
	// contains filtered or unexported fields
}

func (FakeWriterAt) WriteAt 

func (fw FakeWriterAt) WriteAt(p []byte, offset int64) (n int, err error)

type KubernetesConfig 

type KubernetesConfig struct {
	// KubeConfigPath is the path to your kubernetes configuration path
	KubeConfigPath string `json:"kubeConfigPath"`
	// Namespace is the kubernetes namespaces where the pods should be running
	Namespace string `json:"namespace"`
}

type LocalDockerRunner 

type LocalDockerRunner struct {
	// contains filtered or unexported fields
}

LocalDockerRunner is a runner that manually stands up as many docker containers as instances the run job indicates.

It creates a user-defined bridge, to which it attaches a redis service, and all the containers that belong to this test case. It then monitors all test containers, and destroys the setup once all workloads are done.

What we do here is slightly similar to what Docker Compose does, but we can't use the latter because it's a python program and it doesn't expose a network API.

func (*LocalDockerRunner) CollectOutputs 

func (*LocalDockerRunner) CollectOutputs(ctx context.Context, input *api.CollectionInput, ow *rpc.OutputWriter) error

func (*LocalDockerRunner) CompatibleBuilders 

func (*LocalDockerRunner) CompatibleBuilders() []string

func (*LocalDockerRunner) ConfigType 

func (*LocalDockerRunner) ConfigType() reflect.Type

func (*LocalDockerRunner) Healthcheck added in v0.2.0 

func (r *LocalDockerRunner) Healthcheck(ctx context.Context, engine api.Engine, ow *rpc.OutputWriter, fix bool) (*api.HealthcheckReport, error)

func (*LocalDockerRunner) ID 

func (*LocalDockerRunner) ID() string

func (*LocalDockerRunner) Run 

func (r *LocalDockerRunner) Run(ctx context.Context, input *api.RunInput, ow *rpc.OutputWriter) (*api.RunOutput, error)

func (*LocalDockerRunner) TerminateAll added in v0.2.0 

func (*LocalDockerRunner) TerminateAll(ctx context.Context, ow *rpc.OutputWriter) error

This method deletes the testground containers. It does *not* delete any downloaded images or networks. I'll leave a friendly message for how to do a more complete cleanup.

type LocalDockerRunnerConfig 

type LocalDockerRunnerConfig struct {
	// KeepContainers retains test containers even after they exit (default:
	// false).
	KeepContainers bool `toml:"keep_containers"`
	// LogLevel sets the log level in the test containers (default: not set).
	LogLevel string `toml:"log_level"`
	// Unstarted creates the containers without starting them (default: false).
	Unstarted bool `toml:"no_start"`
	// Background avoids tailing the output of containers, and displaying it as
	// log messages (default: false).
	Background bool `toml:"background"`
	// Ulimits that should be applied on this run, in Docker format.
	// See
	// https://docs.docker.com/engine/reference/commandline/run/#set-ulimits-in-container---ulimit
	// (default: ["nofile=1048576:1048576"]).
	Ulimits []string `toml:"ulimits"`
}

LocalDockerRunnerConfig is the configuration object of this runner. Boolean values are expressed in a way that zero value (false) is the default setting.

type LocalExecutableRunner 

type LocalExecutableRunner struct {
	// contains filtered or unexported fields
}

func (*LocalExecutableRunner) Close added in v0.2.0 

func (r *LocalExecutableRunner) Close() error

func (*LocalExecutableRunner) CollectOutputs 

func (*LocalExecutableRunner) CollectOutputs(ctx context.Context, input *api.CollectionInput, ow *rpc.OutputWriter) error

func (*LocalExecutableRunner) CompatibleBuilders 

func (*LocalExecutableRunner) CompatibleBuilders() []string

func (*LocalExecutableRunner) ConfigType 

func (*LocalExecutableRunner) ConfigType() reflect.Type

func (*LocalExecutableRunner) Healthcheck added in v0.2.0 

func (r *LocalExecutableRunner) Healthcheck(ctx context.Context, engine api.Engine, ow *rpc.OutputWriter, fix bool) (*api.HealthcheckReport, error)

func (*LocalExecutableRunner) ID 

func (*LocalExecutableRunner) ID() string

func (*LocalExecutableRunner) Run 

func (r *LocalExecutableRunner) Run(ctx context.Context, input *api.RunInput, ow *rpc.OutputWriter) (*api.RunOutput, error)

func (*LocalExecutableRunner) TerminateAll added in v0.4.0 

func (*LocalExecutableRunner) TerminateAll(ctx context.Context, ow *rpc.OutputWriter) error

type LocalExecutableRunnerCfg 

type LocalExecutableRunnerCfg struct{}

LocalExecutableRunnerCfg is the configuration struct for this runner.

type PrettyPrinter 

type PrettyPrinter struct {
	// contains filtered or unexported fields
}

PrettyPrinter is a logger that sends output to the console.

func NewPrettyPrinter 

func NewPrettyPrinter(ow *rpc.OutputWriter) *PrettyPrinter

NewPrettyPrinter constructs a new console logger.

func (*PrettyPrinter) FailStart 

func (c *PrettyPrinter) FailStart(id string, message interface{})

FailStart should be used to report that an instance failed to start.

func (*PrettyPrinter) Manage 

func (c *PrettyPrinter) Manage(id string, stdout, stderr io.ReadCloser)

Manage should be called on the standard output of all instances. It will send the events to a logger and record whether or not the test passed.

func (*PrettyPrinter) Wait 

func (c *PrettyPrinter) Wait() <-chan error

Wait waits for all running tests to finish and returns an error if any of them failed.

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.