Merge pull request #16 from AustrianDataLAB/dev

feature: create run task implementation (#15)

Author: Sokadyn

.github/workflows/main.yml

@@ -55,8 +55,9 @@ jobs:
         uses: ietf-tools/[email protected]
         with:
           token: ${{ github.token }}
+          patchAll: true
           # fall back to dev because we want to have a valid semver
-          branch: ${{ fromJSON('{"main":"dev"}')[github.ref_name] || github.ref_name }}
+          branch: ${{ fromJSON('{"main":"main"}')[github.ref_name] || github.ref_name }}
           noVersionBumpBehavior: current
 
       - name: Set OPERATOR_VERSION
@@ -198,7 +199,7 @@ jobs:
         with:
           token: ${{ github.token }}
           # calculate the changelog from the last tag to the current dev state
-          fromTag: ${{ fromJSON('{"main":"dev"}')[github.ref_name] || github.ref_name }}
+          fromTag: ${{ fromJSON('{"main":"main"}')[github.ref_name] || github.ref_name }}
           toTag: ${{ env.CURRENT }}
       # Create a new release on GitHub with the semantic OPERATOR_VERSION number
       - name: Create Release

Makefile

@@ -86,11 +86,13 @@ help: ## Display this help.
 ##@ Development
 
 .PHONY: manifests
-manifests: controller-gen ## Generate WebhookConfiguration, ClusterRole and CustomResourceDefinition objects.
+manifests: controller-gen enum-gen ## Generate WebhookConfiguration, ClusterRole and CustomResourceDefinition objects.
+	$(ENUM_GEN) --marshal --ptr -f api/**/enums.go
 	$(CONTROLLER_GEN) rbac:roleName=manager-role crd webhook paths="./..." output:crd:artifacts:config=config/crd/bases
 
 .PHONY: generate
-generate: controller-gen ## Generate code containing DeepCopy, DeepCopyInto, and DeepCopyObject method implementations.
+generate: controller-gen enum-gen ## Generate code containing DeepCopy, DeepCopyInto, and DeepCopyObject method implementations.
+	$(ENUM_GEN) --marshal --ptr -f api/**/enums.go
 	$(CONTROLLER_GEN) object:headerFile="hack/boilerplate.go.txt" paths="./..."
 
 .PHONY: fmt
@@ -173,6 +175,8 @@ $(LOCALBIN):
 ## Tool Binaries
 KUSTOMIZE ?= $(LOCALBIN)/kustomize
 CONTROLLER_GEN ?= $(LOCALBIN)/controller-gen
+ENUM_GEN ?= $(LOCALBIN)/go-enum
+ENUM_GEN_VERSION ?= latest
 ENVTEST ?= $(LOCALBIN)/setup-envtest
 
 ## Tool Versions
@@ -190,6 +194,11 @@ controller-gen: $(CONTROLLER_GEN) ## Download controller-gen locally if necessar
 $(CONTROLLER_GEN): $(LOCALBIN)
 	test -s $(LOCALBIN)/controller-gen || GOBIN=$(LOCALBIN) go install sigs.k8s.io/controller-tools/cmd/controller-gen@$(CONTROLLER_TOOLS_VERSION)
 
+.PHONY: enum-gen
+enum-gen: $(ENUM_GEN) ## Download enum-gen locally if necessary.
+$(ENUM_GEN): $(LOCALBIN)
+	test -s $(LOCALBIN)/go-enum || GOBIN=$(LOCALBIN) go install github.com/abice/go-enum@$(ENUM_GEN_VERSION)
+
 .PHONY: envtest
 envtest: $(ENVTEST) ## Download envtest-setup locally if necessary.
 $(ENVTEST): $(LOCALBIN)

README.md

@@ -8,8 +8,25 @@ ExecDAT is a tool to execute data analysis tasks on Kubernetes. It is designed t
 
 ## Getting Started
 
-You’ll need a Kubernetes cluster to run against. See k3d for a quick way to get a local cluster up and running.
-**Note:** Your controller will automatically use the current context in your kubeconfig file (i.e. whatever cluster `kubectl cluster-info` shows).
+### Prerequisites
+
+* operator-sdk
+* container engine (docker, podman, ...)
+* kubernetes cluster (minikube, k3d, ...)
+
+### Test out the operator
+
+```shell
+make install
+make run
+```
+
+### Run using OLM
+
+```shell
+operator-sdk olm install
+operator-sdk run bundle-upgrade ghcr.io/austriandatalab/execdat-operator-bundle:v0.2.0
+```
 
 ### Running on the cluster
 
@@ -47,33 +64,13 @@ UnDeploy the controller to the cluster:
 make undeploy
 ```
 
-## Contributing
-
-// TODO(user): Add detailed information on how you would like others to contribute to this project
-
 ### How it works
 
 This project aims to follow the Kubernetes [Operator pattern](https://kubernetes.io/docs/concepts/extend-kubernetes/operator/)
 
 It uses [Controllers](https://kubernetes.io/docs/concepts/architecture/controller/)
 which provides a reconcile function responsible for synchronizing resources untile the desired state is reached on the cluster
 
-### Test It Out
-
-1. Install the CRDs into the cluster:
-
-```sh
-make install
-```
-
-2. Run your controller (this will run in the foreground, so switch to a new terminal if you want to leave it running):
-
-```sh
-make run
-```
-
-**NOTE:** You can also run this in one step by running: `make install run`
-
 ### Modifying the API definitions
 
 If you are editing the API definitions, generate the manifests such as CRs or CRDs using:

api/v1alpha1/build_types.go

@@ -36,8 +36,7 @@ type BuildSpec struct {
 
 // BuildStatus defines the observed state of Build
 type BuildStatus struct {
-	// INSERT ADDITIONAL STATUS FIELD - define observed state of cluster
-	// Important: Run "make" to regenerate code after modifying this file
+	CurrentPhase *CurrentPhase `json:"currentPhase"` //TODO: add fields of status and dataurls to kubectl get outputs
 }
 
 //+kubebuilder:object:root=true

api/v1alpha1/builder.go

@@ -5,14 +5,15 @@ import (
 	"k8s.io/utils/pointer"
 )
 
-type PodSpecData struct {
+type BuildPodSpecData struct {
 	INIT_SH    string
 	Dockerfile string
 	ImageName  string
+	ImageTag   string
 }
 
 // SetPodSpec sets the pod spec for the build
-func (build *Build) SetPodSpec(podSpec *kcore.PodSpec, podSpecData PodSpecData) error {
+func (build *Build) SetPodSpec(podSpec *kcore.PodSpec, buildPodSpecData BuildPodSpecData) error {
 
 	podSpec.RestartPolicy = kcore.RestartPolicyNever
 
@@ -27,22 +28,23 @@ func (build *Build) SetPodSpec(podSpec *kcore.PodSpec, podSpecData PodSpecData)
 	}
 	podSpec.Containers = []kcore.Container{
 		{
-			Name:    "buildah",
-			Image:   "quay.io/buildah/stable",
-			Command: []string{"/bin/bash", "-c", "--"},
-			Args:    []string{"trap : TERM INT; echo \"$INIT_SH\" | bash"},
+			Name:            "buildah",
+			Image:           "quay.io/buildah/stable",
+			ImagePullPolicy: kcore.PullIfNotPresent,
+			Command:         []string{"/bin/bash", "-c", "--"},
+			Args:            []string{"trap : TERM INT; echo \"$INIT_SH\" | bash"},
 			Env: []kcore.EnvVar{
-				{Name: "INIT_SH", Value: podSpecData.INIT_SH},
-				{Name: "DOCKERFILE", Value: podSpecData.Dockerfile},
+				{Name: "INIT_SH", Value: buildPodSpecData.INIT_SH},
+				{Name: "DOCKERFILE", Value: buildPodSpecData.Dockerfile},
 				{Name: "BASE_IMAGE", Value: build.Spec.BaseImage},
-				{Name: "IMAGE_NAME", Value: podSpecData.ImageName},
-				// {Name: "IMAGE_TAG", Value: build.Spec.ImageTag},
+				{Name: "IMAGE_NAME", Value: buildPodSpecData.ImageName},
+				{Name: "IMAGE_TAG", Value: buildPodSpecData.ImageTag},
 				// {Name: "IMAGE_REGISTRY", Value: build.Spec.ImageRegistry},
 				// {Name: "IMAGE_REGISTRY_USER", Value: build.Spec.ImageRegistryUser},
 				// {Name: "IMAGE_REGISTRY_PASSWORD", Value: build.Spec.ImageRegistryPassword},
 				// {Name: "IMAGE_REGISTRY_INSECURE", Value: build.Spec.ImageRegistryInsecure},
 				// {Name: "IMAGE_REGISTRY_VERIFY_TLS", Value: build.Spec.ImageRegistryVerifyTLS},
-				{Name: "ENTRYPOINT", Value: build.Spec.SourceCode.Entrypoint},
+				{Name: "ENTRYPOINT", Value: build.Spec.SourceCode.EntryPoint},
 				{Name: "GIT_REPO", Value: build.Spec.SourceCode.URL},
 				{Name: "GIT_BRANCH", Value: build.Spec.SourceCode.Branch},
 				{Name: "BUILD_CMD", Value: build.Spec.SourceCode.BuildCMD},

api/v1alpha1/enums.go

@@ -0,0 +1,5 @@
+package v1alpha1
+
+// swagger:enum CurrentPhase
+// ENUM(pending, building, running, buildComplete, runCompleted, failed)
+type CurrentPhase string

api/v1alpha1/enums_enum.go

@@ -24,17 +24,17 @@ type SourceCodeSpec struct {
 	Dependencies  DependenciesSpec `json:"dependencies,omitempty"`
 	DependencyCMD string           `json:"dependencycmd,omitempty"`
 	BuildCMD      string           `json:"buildcmd,omitempty"`
-	Entrypoint    string           `json:"entrypoint"`
+	EntryPoint    string           `json:"entrypoint"`
 }
 
 type InputDataSpec struct {
-	URL          string `json:"url" description:"URL of the data repo of input data"`
-	Type         string `json:"type" description:"Type of the input data source, e.g. s3, git, http, https, etc."`
 	DataPath     string `json:"datapath" description:"Path to the data directory with input data, has to be a unix path."`
+	URL          string `json:"url" description:"URL of the data repo of input data"`
+	Type         string `json:"type,omitempty" description:"Type of the input data source, e.g. s3, git, http, https, etc."`
 	TransformCMD string `json:"transformcmd,omitempty" description:"Command to transform the input data"`
 }
 
 type OutputDataSpec struct {
-	URL      string `json:"url" description:"URL of the data repo of output data"`
 	DataPath string `json:"datapath" description:"Path to the data directory with output data, has to be a unix path."`
+	URL      string `json:"url,omitempty" description:"URL of the data repo of output data"`
 }

api/v1alpha1/generic_types.go

@@ -29,19 +29,17 @@ type RunSpec struct {
 	// Important: Run "make" to regenerate code after modifying this file
 
 	// Foo is an example field of Run. Edit run_types.go to remove/update
-	Build      BuildSpec      `json:"build"`
-	OutputData OutputDataSpec `json:"outputdata"`
-	//+optional
-	InputData   InputDataSpec `json:"inputdata,omitempty"`
-	Description string        `json:"description,omitempty"`
+	Build       BuildSpec      `json:"build"`
+	OutputData  OutputDataSpec `json:"outputdata,omitempty"`
+	InputData   InputDataSpec  `json:"inputdata,omitempty"`
+	Description string         `json:"description,omitempty"`
 }
 
 // RunStatus defines the observed state of Run
 type RunStatus struct {
 	// INSERT ADDITIONAL STATUS FIELD - define observed state of cluster
 	// Important: Run "make" to regenerate code after modifying this file
-	CurrentPhase string `json:"currentPhase"` //TODO: add fields of status and dataurls to kubectl get outputs
-	Test         string `json:"test"`
+	CurrentPhase *CurrentPhase `json:"currentPhase"` //TODO: add fields of status and dataurls to kubectl get outputs
 }
 
 //+kubebuilder:object:root=true

api/v1alpha1/run_types.go

@@ -0,0 +1,68 @@
+package v1alpha1
+
+import (
+	kcore "k8s.io/api/core/v1"
+	"k8s.io/utils/pointer"
+)
+
+type RunPodSpecData struct {
+	INIT_SH        string
+	ImageName      string
+	ImageTag       string
+	InputDataPath  string
+	OutputDataPath string
+}
+
+// SetPodSpec sets the pod spec for the run
+func (run *Run) SetPodSpec(podSpec *kcore.PodSpec, runPodSpecData RunPodSpecData) error {
+
+	podSpec.RestartPolicy = kcore.RestartPolicyNever
+
+	podSpec.Volumes = []kcore.Volume{
+		{
+			Name: "input",
+			VolumeSource: kcore.VolumeSource{
+				EmptyDir: &kcore.EmptyDirVolumeSource{},
+			},
+		},
+		{
+			Name: "output",
+			VolumeSource: kcore.VolumeSource{
+				EmptyDir: &kcore.EmptyDirVolumeSource{},
+			},
+		},
+	}
+
+	podSpec.Containers = []kcore.Container{
+		{
+			Name:            "buildah",
+			Image:           "harbor.caas-0013.dev.austrianopencloudcommunity.org/execdev/" + runPodSpecData.ImageName + ":" + runPodSpecData.ImageTag,
+			ImagePullPolicy: kcore.PullIfNotPresent,
+			Command:         []string{"/bin/bash", "-c", "--"},
+			Args:            []string{"trap : TERM INT; echo \"$INIT_SH\" | bash"},
+			TTY:             true,
+			Stdin:           true,
+			Env: []kcore.EnvVar{
+				{Name: "INIT_SH", Value: runPodSpecData.INIT_SH},
+				{Name: "MINIO_ENDPOINT", Value: "http://minio.single-minio.svc.cluster.local:9000"},
+				{Name: "MINIO_ACCESS_KEY", Value: "cache-user-1"},
+				{Name: "MINIO_SECRET_KEY", Value: "CACHE_USER_PASS_XYZ_123"},
+				{Name: "BUCKET_NAME", Value: "cache-bucket-1"},
+				{Name: "HOME", Value: "/tmp"},
+			},
+			SecurityContext: &kcore.SecurityContext{
+				RunAsUser:  pointer.Int64(1000),
+				RunAsGroup: pointer.Int64(1000),
+			},
+			VolumeMounts: []kcore.VolumeMount{
+				{Name: "input", MountPath: runPodSpecData.InputDataPath},
+				{Name: "output", MountPath: runPodSpecData.OutputDataPath},
+			},
+		},
+	}
+	podSpec.SecurityContext = &kcore.PodSecurityContext{
+		RunAsUser:  pointer.Int64(1000),
+		RunAsGroup: pointer.Int64(1000),
+	}
+	return nil
+}