Add codeowners.

Completely rebuild the tree generation logic.

.github/workflows/build-test-release.yaml

@@ -100,6 +100,7 @@ jobs:
     name: "Building docker image"
     needs: [ prepare, test ]
     runs-on: ubuntu-20.04
+    if: needs.prepare.outputs.CHECK_IF_MASTER_BRANCH == 'true'
     steps:
       - name: Checkout repo
         uses: actions/checkout@v3

.github/workflows/codeql.yml

@@ -0,0 +1,76 @@
+# For most projects, this workflow file will not need changing; you simply need
+# to commit it to your repository.
+#
+# You may wish to alter this file to override the set of languages analyzed,
+# or to provide custom queries or build logic.
+#
+# ******** NOTE ********
+# We have attempted to detect the languages in your repository. Please check
+# the `language` matrix defined below to confirm you have the correct set of
+# supported CodeQL languages.
+#
+name: "CodeQL"
+
+on:
+  push:
+    branches: [ "main" ]
+  pull_request:
+    # The branches below must be a subset of the branches above
+    branches: [ "main" ]
+  schedule:
+    - cron: '22 9 * * 0'
+
+jobs:
+  analyze:
+    name: Analyze
+    runs-on: ubuntu-latest
+    permissions:
+      actions: read
+      contents: read
+      security-events: write
+
+    strategy:
+      fail-fast: false
+      matrix:
+        language: [ 'go' ]
+        # CodeQL supports [ 'cpp', 'csharp', 'go', 'java', 'javascript', 'python', 'ruby' ]
+        # Use only 'java' to analyze code written in Java, Kotlin or both
+        # Use only 'javascript' to analyze code written in JavaScript, TypeScript or both
+        # Learn more about CodeQL language support at https://aka.ms/codeql-docs/language-support
+
+    steps:
+    - name: Checkout repository
+      uses: actions/checkout@v3
+
+    # Initializes the CodeQL tools for scanning.
+    - name: Initialize CodeQL
+      uses: github/codeql-action/init@v2
+      with:
+        languages: ${{ matrix.language }}
+        # If you wish to specify custom queries, you can do so here or in a config file.
+        # By default, queries listed here will override any specified in a config file.
+        # Prefix the list here with "+" to use these queries and those in the config file.
+
+        # Details on CodeQL's query packs refer to : https://docs.github.com/en/code-security/code-scanning/automatically-scanning-your-code-for-vulnerabilities-and-errors/configuring-code-scanning#using-queries-in-ql-packs
+        # queries: security-extended,security-and-quality
+
+
+    # Autobuild attempts to build any compiled languages  (C/C++, C#, Go, or Java).
+    # If this step fails, then you should remove it and run the build manually (see below)
+    - name: Autobuild
+      uses: github/codeql-action/autobuild@v2
+
+    # ℹ️ Command-line programs to run using the OS shell.
+    # 📚 See https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#jobsjob_idstepsrun
+
+    #   If the Autobuild fails above, remove it and uncomment the following three lines.
+    #   modify them (or add more) to build your code if your project, please refer to the EXAMPLE below for guidance.
+
+    # - run: |
+    #     echo "Run, Build Application using script"
+    #     ./location_of_script_within_repo/buildscript.sh
+
+    - name: Perform CodeQL Analysis
+      uses: github/codeql-action/analyze@v2
+      with:
+        category: "/language:${{matrix.language}}"

CODEOWNERS

@@ -0,0 +1 @@
+* lukaszraczylo

README.md

@@ -28,7 +28,6 @@ metadata:
   labels:
   name: managedjob-sample
 spec:
-  # Globally defined parameters and environment variables
   retries: 3
   params:
     env:
@@ -40,9 +39,7 @@ spec:
   # Job groups definitions
   groups:
     - name: "first-group"
-      # Group will run in parallel with other defined groups
       parallel: true
-      # Group specific parameters
       params:
         env:
           - name: "FEE"
@@ -63,6 +60,11 @@ spec:
           args:
             - "sleep"
             - "10"
+        - name: "second-half-job"
+          image: "busybox"
+          args:
+            - "sleep"
+            - "10"
 
     - name: "second-group"
       parallel: true
@@ -79,9 +81,44 @@ spec:
           args:
             - "sleep"
             - "10"
+          parallel: false
+
+    - name: "third-group"
+      parallel: false
+      jobs:
+        - name: "fifth-job"
+          image: "busybox"
+          args:
+            - "echo"
+            - "Hello world!"
           parallel: true
 ```
 
+### How does it look in practice?
+
+```yaml
+managedjob-sample
+├── first-group
+│   ├── first-job
+│   ├── second-job
+│   │   └── Depends on: managedjob-sample-first-group-first-job
+│   └── second-half-job
+│       ├── Depends on: managedjob-sample-first-group-first-job
+│       └── Depends on: managedjob-sample-first-group-second-job
+├── second-group
+│   ├── third-job
+│   └── fourth-job
+│       └── Depends on: managedjob-sample-second-group-third-job
+└── third-group
+    ├── fifth-job
+    ├── Depends on group: first-group
+    └── Depends on group: second-group
+```
+
+If dependency exists on the group level - the group will not be executed until all of remaining groups have finished successfuly.
+If dependency exists on the job level - the job will not be executed until all of remaining jobs have finished successfuly.
+Remember that **ORDER matters**.
+
 ### Things to remember
 
 Parameters **params** are always merged downwards to DRY your definitions.
@@ -108,6 +145,30 @@ In this case - result for the first job will look like this:
         status: succeeded
 ```
 
+### Kustomization and references
+
+In case of any issues with `configmapGenerator` or `secretGenerator`, please add following to your `kustomization.yaml`:
+
+```yaml
+configurations:
+  - crd-name-reference.yaml
+```
+
+Then you can create `crd-name-reference.yaml` file with following content:
+
+```yaml
+---
+nameReference:
+  - kind: 'ConfigMap'
+    fieldSpecs:
+      - kind: 'ManagedJob'
+        path: 'spec/params/fromEnv[]/configMapRef/name'
+      - kind: 'ManagedJob'
+        path: 'spec/params/env[]/configMapRef/name'
+```
+
+This will instruct kustomize to replace all references to configmaps with their names if they are managed by generators.
+
 ### Running on the cluster
 
 #### Manual installation

api/v1beta1/managedjob_types.go

@@ -24,10 +24,8 @@ import (
 type ManagedJobDependencies struct {
 	// +kubebuilder:validation:Optional
 	// +kubebuilder:default=""
-	Group string `json:"group"`
-	// +kubebuilder:validation:Optional
-	// +kubebuilder:default=""
-	Job string `json:"job"`
+	Name   string `json:"name"`
+	Status string `json:"status"`
 }
 
 type ManagedJobDefinition struct {
@@ -48,8 +46,9 @@ type ManagedJobDefinition struct {
 	// +kubebuilder:validation:Optional
 	// +kubebuilder:default=pending
 	Status string `json:"status"`
+	// +kubebuilder:validation:Optional
 	// +optional
-	Dependencies ManagedJobDependencies `json:"dependencies"`
+	Dependencies []*ManagedJobDependencies `json:"dependencies"`
 	// +optional
 	CompiledParams ManagedJobParameters `json:"compiledParams"`
 }
@@ -68,6 +67,9 @@ type ManagedJobGroup struct {
 	// +kubebuilder:validation:Optional
 	Params ManagedJobParameters `json:"params"`
 	// +kubebuilder:validation:Optional
+	// +optional
+	Dependencies []*ManagedJobDependencies `json:"dependencies"`
+	// +kubebuilder:validation:Optional
 	// +kubebuilder:default=pending
 	Status string `json:"status"`
 }
@@ -105,9 +107,6 @@ type ManagedJobSpec struct {
 	Groups []*ManagedJobGroup `json:"groups"`
 	// +kubebuilder:validation:Optional
 	Params ManagedJobParameters `json:"params"`
-	// +kubebuilder:validation:Optional
-	// +kubebuilder:default=pending
-	Status string `json:"status"`
 }
 
 // +kubebuilder:object:root=true

api/v1beta1/zz_generated.deepcopy.go

@@ -61,7 +61,17 @@ func (in *ManagedJobDefinition) DeepCopyInto(out *ManagedJobDefinition) {
 		copy(*out, *in)
 	}
 	in.Params.DeepCopyInto(&out.Params)
-	out.Dependencies = in.Dependencies
+	if in.Dependencies != nil {
+		in, out := &in.Dependencies, &out.Dependencies
+		*out = make([]*ManagedJobDependencies, len(*in))
+		for i := range *in {
+			if (*in)[i] != nil {
+				in, out := &(*in)[i], &(*out)[i]
+				*out = new(ManagedJobDependencies)
+				**out = **in
+			}
+		}
+	}
 	in.CompiledParams.DeepCopyInto(&out.CompiledParams)
 }
 
@@ -105,6 +115,17 @@ func (in *ManagedJobGroup) DeepCopyInto(out *ManagedJobGroup) {
 		}
 	}
 	in.Params.DeepCopyInto(&out.Params)
+	if in.Dependencies != nil {
+		in, out := &in.Dependencies, &out.Dependencies
+		*out = make([]*ManagedJobDependencies, len(*in))
+		for i := range *in {
+			if (*in)[i] != nil {
+				in, out := &(*in)[i], &(*out)[i]
+				*out = new(ManagedJobDependencies)
+				**out = **in
+			}
+		}
+	}
 }
 
 // DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new ManagedJobGroup.

charts/jobs-manager-operator/Chart.yaml

@@ -13,12 +13,12 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.0.2
+version: 0.0.19
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.
 # It is recommended to use it with quotes.
-appVersion: "0.0.2"
+appVersion: "0.0.19"
 keywords:
   - operator
   - jobs

charts/jobs-manager-operator/templates/managedjob-crd.yaml

@@ -38,6 +38,18 @@ spec:
               groups:
                 items:
                   properties:
+                    dependencies:
+                      items:
+                        properties:
+                          name:
+                            default: ""
+                            type: string
+                          status:
+                            type: string
+                        required:
+                        - status
+                        type: object
+                      type: array
                     jobs:
                       items:
                         properties:
@@ -2100,14 +2112,17 @@ spec:
                                 type: array
                             type: object
                           dependencies:
-                            properties:
-                              group:
-                                default: ""
-                                type: string
-                              job:
-                                default: ""
-                                type: string
-                            type: object
+                            items:
+                              properties:
+                                name:
+                                  default: ""
+                                  type: string
+                                status:
+                                  type: string
+                              required:
+                              - status
+                              type: object
+                            type: array
                           image:
                             minLength: 5
                             type: string
@@ -7912,9 +7927,6 @@ spec:
                 default: 1
                 minimum: 1
                 type: integer
-              status:
-                default: pending
-                type: string
             required:
             - groups
             - retries

charts/jobs-manager-operator/templates/manager-rbac.yaml

@@ -44,6 +44,15 @@ rules:
   - patch
   - update
   - watch
+# Add access to events
+- apiGroups:
+  - ""
+  resources:
+  - events
+  verbs:
+  - create
+  - patch
+  - update
 ---
 apiVersion: rbac.authorization.k8s.io/v1
 kind: ClusterRoleBinding

charts/jobs-manager-operator/values.yaml

@@ -12,8 +12,8 @@ controllerManager:
         memory: 64Mi
   manager:
     image:
-      repository: ghcr.io/lukaszraczylo/jobsmanager
-      tag: 0.0.2
+      repository: ghcr.io/lukaszraczylo/jobs-manager-operator
+      tag: 0.0.19
     resources:
       limits:
         cpu: 500m

config/crd/bases/jobsmanager.raczylo.com_managedjobs.yaml

@@ -38,6 +38,18 @@ spec:
               groups:
                 items:
                   properties:
+                    dependencies:
+                      items:
+                        properties:
+                          name:
+                            default: ""
+                            type: string
+                          status:
+                            type: string
+                        required:
+                        - status
+                        type: object
+                      type: array
                     jobs:
                       items:
                         properties:
@@ -2170,14 +2182,17 @@ spec:
                                 type: array
                             type: object
                           dependencies:
-                            properties:
-                              group:
-                                default: ""
-                                type: string
-                              job:
-                                default: ""
-                                type: string
-                            type: object
+                            items:
+                              properties:
+                                name:
+                                  default: ""
+                                  type: string
+                                status:
+                                  type: string
+                              required:
+                              - status
+                              type: object
+                            type: array
                           image:
                             minLength: 5
                             type: string
@@ -8135,9 +8150,6 @@ spec:
                 default: 1
                 minimum: 1
                 type: integer
-              status:
-                default: pending
-                type: string
             required:
             - groups
             - retries

config/samples/jobsmanager_v1beta1_managedjob.yaml

@@ -41,6 +41,11 @@ spec:
           args:
             - "sleep"
             - "10"
+        - name: "second-half-job"
+          image: "busybox"
+          args:
+            - "sleep"
+            - "10"
 
     - name: "second-group"
       parallel: true
@@ -57,4 +62,14 @@ spec:
           args:
             - "sleep"
             - "10"
+          parallel: false
+
+    - name: "third-group"
+      parallel: false
+      jobs:
+        - name: "fifth-job"
+          image: "busybox"
+          args:
+            - "echo"
+            - "Hello world!"
           parallel: true

controllers/crd_calc.go

@@ -1,71 +0,0 @@
-package controllers
-
-import (
-	corev1 "k8s.io/api/core/v1"
-)
-
-func (cp *connPackage) checkGroupsStatus() {
-	groupsTotal := len(cp.mj.Spec.Groups)
-	totalJobs := 0
-	completedJobs := 0
-	didAnyJobAbort := false
-	changePresent := false
-	groupsCompleted := 0
-
-	// Check if all groups have completed and set ManagedJob status to "succeeded" if so.
-	if cp.mj.Spec.Status != ExecutionStatusSucceeded && groupsTotal > 0 {
-		for _, group := range cp.mj.Spec.Groups {
-			groupJobsTotal := len(group.Jobs)
-			groupJobsCompleted := 0
-
-			for _, job := range group.Jobs {
-				if job.Status == ExecutionStatusSucceeded {
-					groupJobsCompleted++
-					completedJobs++
-				}
-				if job.Status == ExecutionStatusFailed || job.Status == ExecutionStatusAborted {
-					didAnyJobAbort = true
-				}
-				totalJobs++
-			}
-
-			if groupJobsTotal == groupJobsCompleted {
-				// All the jobs in the group are completed.
-				if group.Status != ExecutionStatusSucceeded {
-					group.Status = ExecutionStatusSucceeded
-					changePresent = true
-				}
-				groupsCompleted++
-			}
-		}
-	}
-
-	if groupsTotal == groupsCompleted && cp.mj.Spec.Status != ExecutionStatusSucceeded && cp.mj.Status != ExecutionStatusSucceeded {
-		cp.mj.Spec.Status = ExecutionStatusSucceeded
-		changePresent = true
-		cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Completed", "All jobs completed")
-	}
-
-	// Update status if any job aborted.
-	if didAnyJobAbort && cp.mj.Spec.Status != ExecutionStatusFailed {
-		cp.mj.Spec.Status = ExecutionStatusFailed
-		changePresent = true
-		cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Aborted", "One of the jobs aborted")
-	}
-
-	// Update status to "running" if not already set.
-	// if cp.mj.Spec.Status != ExecutionStatusRunning && cp.mj.Spec.Status != ExecutionStatusSucceeded {
-	// 	cp.mj.Spec.Status = ExecutionStatusRunning
-	// 	cp.mj.Status = ExecutionStatusRunning
-	// 	changePresent = true
-	// }
-
-	// Check if the ManagedJob status has changed.
-	statusChanged := cp.mj.Spec.Status != cp.mj.Status
-
-	// Update status and send event if it has changed.
-	if statusChanged || changePresent {
-		cp.updateCRDStatusDirectly()
-		// cp.r.Client.Status().Update(cp.ctx, cp.mj)
-	}
-}

controllers/crd_dependency_tree.go

@@ -0,0 +1,160 @@
+package controllers
+
+import (
+	"strings"
+
+	"github.com/lukaszraczylo/pandati"
+	jobsmanagerv1beta1 "raczylo.com/jobs-manager-operator/api/v1beta1"
+)
+
+const (
+	newLine      = "\n"
+	emptySpace   = "    "
+	middleItem   = "├── "
+	continueItem = "│   "
+	lastItem     = "└── "
+)
+
+func New(text string) Tree {
+	return &tree{
+		text:  text,
+		items: []Tree{},
+	}
+}
+
+// Add adds a node to the tree
+func (t *tree) Add(text string) Tree {
+	n := New(text)
+	t.items = append(t.items, n)
+	return n
+}
+
+// AddTree adds a tree as an item
+func (t *tree) AddTree(tree Tree) {
+	t.items = append(t.items, tree)
+}
+
+// Text returns the node's value
+func (t *tree) Text() string {
+	return t.text
+}
+
+// Items returns all items in the tree
+func (t *tree) Items() []Tree {
+	return t.items
+}
+
+func (t *tree) Print() string {
+	p := &printer{}
+	return p.Print(t)
+}
+
+func (p *printer) Print(t Tree) string {
+	return t.Text() + newLine + p.printItems(t.Items(), []bool{})
+}
+
+func (p *printer) printText(text string, spaces []bool, last bool) string {
+	var result string
+	for _, space := range spaces {
+		if space {
+			result += emptySpace
+		} else {
+			result += continueItem
+		}
+	}
+
+	indicator := middleItem
+	if last {
+		indicator = lastItem
+	}
+
+	var out string
+	lines := strings.Split(text, "\n")
+	for i := range lines {
+		text := lines[i]
+		if i == 0 {
+			out += result + indicator + text + newLine
+			continue
+		}
+		if last {
+			indicator = emptySpace
+		} else {
+			indicator = continueItem
+		}
+		out += result + indicator + text + newLine
+	}
+
+	return out
+}
+
+func (p *printer) printItems(t []Tree, spaces []bool) string {
+	var result string
+	for i, f := range t {
+		last := i == len(t)-1
+		result += p.printText(f.Text(), spaces, last)
+		if len(f.Items()) > 0 {
+			spacesChild := append(spaces, last)
+			result += p.printItems(f.Items(), spacesChild)
+		}
+	}
+	return result
+}
+
+func (cp *connPackage) checkIfPresentInDependencies(currentDependencies []*jobsmanagerv1beta1.ManagedJobDependencies, dependencyName string) bool {
+	for _, dependency := range currentDependencies {
+		if dependency.Name == dependencyName {
+			return true
+		}
+	}
+	return false
+}
+
+func (cp *connPackage) generateDependencyTree() {
+	// First pass - initialize the tree and get all the gathered jobs
+	originalMainJobDefinition := cp.mj.DeepCopy()
+
+	mainTree := New(cp.mj.Name)
+	for _, group := range cp.mj.Spec.Groups {
+		groupTree := mainTree.Add(group.Name)
+		for _, job := range group.Jobs {
+			jobTree := groupTree.Add(job.Name)
+			if job.Parallel {
+				continue
+			} else {
+				// get the groupTree items before this job and add them as dependencies
+				for _, jobTreePrevious := range groupTree.Items() {
+					if jobTreePrevious.Text() == job.Name {
+						break
+					}
+					generatedJobName := jobNameGenerator(cp.mj.Name, group.Name, jobTreePrevious.Text())
+					jobTree.Add("Depends on: " + generatedJobName)
+					if !cp.checkIfPresentInDependencies(job.Dependencies, generatedJobName) {
+						job.Dependencies = append(job.Dependencies, &jobsmanagerv1beta1.ManagedJobDependencies{Name: generatedJobName, Status: ExecutionStatusPending})
+						job.CompiledParams = cp.compileParameters(cp.mj.Spec.Params, group.Params, job.Params)
+					}
+				}
+			}
+		}
+		if group.Parallel {
+			continue
+		} else {
+			// get the mainTree items before this group and add them as dependencies
+			for _, groupTreePrevious := range mainTree.Items() {
+				if groupTreePrevious.Text() == group.Name {
+					break
+				}
+				groupTree.Add("Depends on group: " + groupTreePrevious.Text())
+				if !cp.checkIfPresentInDependencies(group.Dependencies, groupTreePrevious.Text()) {
+					group.Dependencies = append(group.Dependencies, &jobsmanagerv1beta1.ManagedJobDependencies{Name: groupTreePrevious.Text(), Status: ExecutionStatusPending})
+				}
+			}
+		}
+	}
+
+	_, theSame, _ := pandati.CompareStructsReplaced(originalMainJobDefinition, cp.mj)
+	if !theSame {
+		cp.updateCRDStatusDirectly()
+	}
+	// fmt.Print(mainTree.Print())
+	// fmt.Printf("Dependency tree: %# v", pretty.Formatter(mainTree))
+}

controllers/crd_scrapper.go

@@ -1,6 +1,8 @@
 package controllers
 
 import (
+	"strings"
+
 	"github.com/lukaszraczylo/pandati"
 	kbatch "k8s.io/api/batch/v1"
 	corev1 "k8s.io/api/core/v1"
@@ -34,9 +36,9 @@ func (cp *connPackage) compileParameters(params ...jobsmanagerv1beta1.ManagedJob
 			if params.Env != nil {
 				cparams.Env = append(cparams.Env, params.Env...)
 			}
-			// if params.Volumes != nil {
-			// 	cparams.Volumes = append(cparams.Volumes, params.Volumes...)
-			// }
+			if params.Volumes != nil {
+				cparams.Volumes = append(cparams.Volumes, params.Volumes...)
+			}
 			if params.VolumeMounts != nil {
 				cparams.VolumeMounts = append(cparams.VolumeMounts, params.VolumeMounts...)
 			}
@@ -59,135 +61,166 @@ func (cp *connPackage) compileParameters(params ...jobsmanagerv1beta1.ManagedJob
 	return cparams
 }
 
-type previousJobsAndGroups struct {
-	GroupID string
-	JobID   string
-}
-
-func (cp *connPackage) buildJobsDependencyTree() error {
-	changePending := false
-	for i, group := range cp.mj.Spec.Groups {
-		var group_previous string
-		if group.Parallel {
-			group_previous = group.Name
-		} else {
-			if i > 0 {
-				group_previous = cp.mj.Spec.Groups[i-1].Name
-			} else {
-				group_previous = group.Name
-			}
-		}
-		for j, job := range group.Jobs {
-			if job.Dependencies.Group == "" || job.Dependencies.Job == "" {
-				changePending = true
-			}
-			job.Dependencies.Group = group_previous
-			if job.Parallel {
-				job.Dependencies.Job = job.Name
-			} else {
-				if j > 0 {
-					job.Dependencies.Job = group.Jobs[j-1].Name
-				} else {
-					job.Dependencies.Job = job.Name
+func (cp *connPackage) updateDependentJobs(completedJob string, jobStatus string) {
+	for _, group := range cp.mj.Spec.Groups {
+		for _, job := range group.Jobs {
+			for _, dependency := range job.Dependencies {
+				if dependency.Name == completedJob && dependency.Status != jobStatus {
+					dependency.Status = jobStatus
 				}
 			}
 		}
 	}
-	if changePending {
-		cp.updateCRDStatusDirectly()
-	}
-	return nil
 }
 
-func (cp *connPackage) checkJobStatus() {
+func (cp *connPackage) updateDependentGroups(completedGroup string, jobStatus string) {
+	for _, group := range cp.mj.Spec.Groups {
+		for _, dependency := range group.Dependencies {
+			if dependency.Name == completedGroup && dependency.Status != jobStatus {
+				dependency.Status = jobStatus
+			}
+		}
+	}
+}
+
+func (cp *connPackage) checkRunningJobsStatus() {
 	var childJobs kbatch.JobList
 	labelSelector := labels.SelectorFromSet(labels.Set{
 		"jobmanager.raczylo.com/workflow-name": cp.mj.Name,
 	})
 	listOptions := &client.ListOptions{LabelSelector: labelSelector, Namespace: cp.mj.Namespace}
-
 	err := cp.r.Client.List(cp.ctx, &childJobs, listOptions)
 	if err != nil {
-		// log.Log.Error(err, "unable to list child jobs")
+		log.Log.Info("Unable to list child jobs", "error", err.Error())
 		return
 	}
-	changePresent := false
+
 	for _, childJob := range childJobs.Items {
 		for _, group := range cp.mj.Spec.Groups {
 			for _, job := range group.Jobs {
 				generatedJobName := jobNameGenerator(cp.mj.Name, group.Name, job.Name)
 				if childJob.Name == generatedJobName {
 					if childJob.Status.Succeeded > 0 && job.Status != ExecutionStatusSucceeded {
+						cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Completed", "Job %s completed [prev: %s]", childJob.Name, job.Status)
 						job.Status = ExecutionStatusSucceeded
-						changePresent = true
-						cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Completed", "Job %s completed", childJob.Name)
 					} else if childJob.Status.Failed > 0 && job.Status != ExecutionStatusFailed {
+						cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeWarning, "Failed", "Job %s failed [prev: %s]", childJob.Name, job.Status)
 						job.Status = ExecutionStatusFailed
-						changePresent = true
-						cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Failed", "Job %s failed", childJob.Name)
 					} else if childJob.Status.Active > 0 && job.Status != ExecutionStatusRunning {
-						changePresent = true
+						cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Running", "Job %s running [prev: %s]", childJob.Name, job.Status)
 						job.Status = ExecutionStatusRunning
 					}
+					cp.updateDependentJobs(generatedJobName, job.Status)
+					continue
 				}
 			}
-
 		}
 	}
-	if changePresent {
-		cp.updateCRDStatusDirectly()
-	}
 }
 
 func (cp *connPackage) runPendingJobs() {
-	copyMJ := cp.mj.DeepCopy()
-	changePresent := false
+	// originalMainJobDefinition := cp.mj.DeepCopy()
 	for _, group := range cp.mj.Spec.Groups {
+		run_group := false
+
+		groupJobsCompleted := 0
 		for _, job := range group.Jobs {
-			if job.Status == ExecutionStatusPending {
-				if job.Dependencies.Group == group.Name && job.Dependencies.Job == job.Name {
-					job.CompiledParams = cp.compileParameters(job.Params, group.Params, cp.mj.Spec.Params)
-					err := cp.executeJob(job, group)
-					if err != nil {
-						// log.Log.Info("Unable to execute job", "group", group.Name, "job", job.Name, "error", err)
-						continue
+			if job.Status == ExecutionStatusSucceeded {
+				groupJobsCompleted++
+			}
+		}
+		if groupJobsCompleted == len(group.Jobs) {
+			group.Status = ExecutionStatusSucceeded
+			cp.updateDependentGroups(group.Name, group.Status)
+			continue
+		}
+
+		approvedStatuses := []string{ExecutionStatusSucceeded, ExecutionStatusFailed, ExecutionStatusAborted}
+		if pandati.ExistsInSlice(approvedStatuses, group.Status) {
+			cp.updateDependentGroups(group.Name, group.Status)
+		}
+
+		approvedStatuses = []string{ExecutionStatusPending, ExecutionStatusRunning}
+		if pandati.ExistsInSlice(approvedStatuses, group.Status) {
+			if len(group.Dependencies) > 0 {
+				groupsCompleted := 0
+				for _, group_dependency := range group.Dependencies {
+					if group_dependency.Status == ExecutionStatusSucceeded {
+						groupsCompleted++
 					}
-					job.Status = ExecutionStatusRunning
-					changePresent = true
-					continue
-				} else {
-					for _, group2 := range copyMJ.Spec.Groups {
-						for _, job2 := range group2.Jobs {
-							if job2.Name == job.Dependencies.Job && group2.Name == job.Dependencies.Group {
-								switch job2.Status {
-								case ExecutionStatusSucceeded:
-									job.CompiledParams = cp.compileParameters(job.Params, group.Params, cp.mj.Spec.Params)
-									err := cp.executeJob(job, group)
-									if err != nil {
-										log.Log.Info("Unable to execute job", "group", group.Name, "job", job.Name, "error", err)
-										continue
-									}
-									job.Status = ExecutionStatusRunning
-									changePresent = true
-								case ExecutionStatusFailed:
+					if group_dependency.Status == ExecutionStatusFailed {
+						group.Status = ExecutionStatusAborted
+						cp.updateDependentGroups(group.Name, ExecutionStatusFailed)
+					}
+				}
+				if groupsCompleted == len(group.Dependencies) {
+					run_group = true
+				}
+			} else {
+				run_group = true
+			}
+
+			if !run_group {
+				// fmt.Println("Group "+group.Name+" is not running as dependencies were not met", group.Dependencies)
+				continue // not running the group as dependencies were not met
+			} else {
+				group.Status = ExecutionStatusRunning
+				cp.updateDependentGroups(group.Name, ExecutionStatusRunning)
+
+				for _, job := range group.Jobs {
+					run_job := false
+					if job.Status == ExecutionStatusPending {
+						if len(job.Dependencies) > 0 {
+							jobsCompleted := 0
+							for _, job_dependency := range job.Dependencies {
+								if job_dependency.Status == ExecutionStatusSucceeded {
+									jobsCompleted++
+								}
+								if job_dependency.Status == ExecutionStatusFailed {
 									job.Status = ExecutionStatusAborted
-									changePresent = true
+									cp.updateDependentJobs(job.Name, ExecutionStatusFailed)
 								}
 							}
+							if jobsCompleted == len(job.Dependencies) {
+								run_job = true
+							}
+						} else {
+							run_job = true
+						}
+					}
+
+					if !run_job {
+						continue // job is not ready as dependencies were not met
+					} else {
+						approvedStatuses = []string{ExecutionStatusRunning, ExecutionStatusFailed, ExecutionStatusAborted}
+						if !pandati.ExistsInSlice(approvedStatuses, job.Status) {
+							err := cp.executeJob(job, group)
+							if err != nil {
+								log.Log.Info("Unable to execute job", "error", err.Error())
+								if !strings.Contains(err.Error(), "exists") {
+									job.Status = ExecutionStatusFailed
+									group.Status = ExecutionStatusFailed
+									cp.updateDependentJobs(job.Name, ExecutionStatusFailed)
+									cp.updateDependentGroups(group.Name, ExecutionStatusFailed)
+									cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeWarning, "Failed", "Job %s from group %s failed", job.Name, group.Name)
+								}
+								return
+							}
+							job.Status = ExecutionStatusRunning
+							cp.updateDependentJobs(job.Name, ExecutionStatusRunning)
+							cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Running", "Job %s from group %s running", job.Name, group.Name)
 						}
 					}
 				}
 			}
+
+			// fmt.Println("Running group: ", group.Name, " with status: ", group.Status, " accepted: ", run_group)
 		}
 	}
-	if changePresent {
-		cp.updateCRDStatusDirectly()
-	}
 }
 
 func (cp *connPackage) executeJob(j *jobsmanagerv1beta1.ManagedJobDefinition, g *jobsmanagerv1beta1.ManagedJobGroup) (err error) {
 	generatedJobName := jobNameGenerator(cp.mj.Name, g.Name, j.Name)
-
 	convertRetries := func(retries int) *int32 {
 		if retries == 0 {
 			return nil
@@ -252,10 +285,34 @@ func (cp *connPackage) executeJob(j *jobsmanagerv1beta1.ManagedJobDefinition, g
 
 	err = cp.r.Client.Create(cp.ctx, &job_handler)
 	if err != nil || pandati.IsZero(job_handler) {
-		log.Log.Error(err, "Unable to create job", "job", job_handler.Name)
 		return err
 	}
 
 	cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Created", "Created job %s", job_handler.Name)
 	return nil
 }
+
+func (cp *connPackage) checkOverallStatus() {
+	groupsCompleted := 0
+	negativeStatuses := []string{ExecutionStatusFailed, ExecutionStatusAborted}
+	for _, group := range cp.mj.Spec.Groups {
+		if group.Status == ExecutionStatusSucceeded {
+			groupsCompleted++
+		} else if pandati.ExistsInSlice(negativeStatuses, group.Status) {
+			cp.mj.Status = ExecutionStatusFailed
+			cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeWarning, "Failure", "Run failed in group %s", group.Name)
+		} else {
+			continue
+		}
+	}
+
+	if groupsCompleted == len(cp.mj.Spec.Groups) {
+		if cp.mj.Status != ExecutionStatusSucceeded {
+			cp.r.Recorder.Eventf(cp.mj, corev1.EventTypeNormal, "Success", "Run completed successfuly")
+		}
+		cp.mj.Status = ExecutionStatusSucceeded
+	} else {
+		cp.mj.Status = ExecutionStatusRunning
+	}
+	cp.r.Status().Update(cp.ctx, cp.mj)
+}

controllers/definitions.go

@@ -1,8 +1,6 @@
 package controllers
 
 // +kubebuilder:validation:Enum=Allow;Forbid;Replace
-type ExecutionStatus string
-
 const (
 	ExecutionStatusPending   string = "pending"
 	ExecutionStatusRunning   string = "running"
@@ -15,3 +13,29 @@ const (
 var (
 	jobOwnerKey = ".metadata.controller"
 )
+
+type (
+	ExecutionStatus string
+
+	tree struct {
+		text  string
+		items []Tree
+	}
+
+	// Tree is tree interface
+	Tree interface {
+		Add(text string) Tree
+		AddTree(tree Tree)
+		Items() []Tree
+		Text() string
+		Print() string
+	}
+
+	printer struct {
+	}
+
+	// Printer is printer interface
+	Printer interface {
+		Print(Tree) string
+	}
+)

controllers/helpers.go

@@ -2,19 +2,13 @@ package controllers
 
 import (
 	"context"
-	"encoding/json"
-	"fmt"
 	"strings"
 	"sync"
 
-	"github.com/kr/pretty"
-	"github.com/lukaszraczylo/pandati"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
-	"k8s.io/apimachinery/pkg/types"
 	"raczylo.com/jobs-manager-operator/api/v1beta1"
 	jobsmanagerv1beta1 "raczylo.com/jobs-manager-operator/api/v1beta1"
 	ctrl "sigs.k8s.io/controller-runtime"
-	"sigs.k8s.io/controller-runtime/pkg/client"
 	"sigs.k8s.io/controller-runtime/pkg/log"
 )
 
@@ -30,11 +24,12 @@ type jobStatusUpdate struct {
 }
 
 type connPackage struct {
-	r   *ManagedJobReconciler
-	ctx context.Context
-	req ctrl.Request
-	mtx sync.Mutex
-	mj  *jobsmanagerv1beta1.ManagedJob
+	r              *ManagedJobReconciler
+	ctx            context.Context
+	req            ctrl.Request
+	mtx            sync.Mutex
+	mj             *jobsmanagerv1beta1.ManagedJob
+	dependencyTree Tree
 }
 
 func (cp *connPackage) getOwnerReference() (metav1.OwnerReference, error) {
@@ -53,65 +48,8 @@ func (cp *connPackage) getOwnerReference() (metav1.OwnerReference, error) {
 	}, nil
 }
 
-type patchStringValue struct {
-	Op    string `json:"op"`
-	Path  string `json:"path"`
-	Value string `json:"value"`
-}
-
-func (cp *connPackage) getLatestMainJobAndPatch(proposedChanges *jobsmanagerv1beta1.ManagedJob) (bool, error) {
-	mj := jobsmanagerv1beta1.ManagedJob{}
-	err := cp.r.Client.Get(cp.ctx, cp.req.NamespacedName, &mj)
-	if err != nil {
-		return false, err
-	}
-	diff, diffIdentical, _ := pandati.CompareStructsReplaced(mj, proposedChanges)
-	if !diffIdentical {
-		if err != nil {
-			log.Log.Error(err, "Unable to marshal ManagedJob")
-		}
-		var payloadBytesArr []patchStringValue
-		for _, d := range diff {
-			if strings.HasPrefix(d.Key, "/spec") {
-				operation := "replace"
-				if d.OldValue == "" || d.OldValue == nil {
-					operation = "add"
-				}
-
-				payload := patchStringValue{
-					Op:    operation,
-					Path:  d.Key,
-					Value: d.Value.(string),
-				}
-				payloadBytesArr = append(payloadBytesArr, payload)
-			}
-		}
-		patchPayload, err := json.Marshal(payloadBytesArr)
-		if err != nil {
-			log.Log.Error(err, "Unable to marshal ManagedJob")
-		}
-		fmt.Printf("PatchPayload: %# v", pretty.Formatter(fmt.Sprintf("%s", patchPayload)))
-		kubepath := client.RawPatch(types.JSONPatchType, patchPayload)
-		err = cp.r.Client.Patch(cp.ctx, &mj, kubepath)
-		// if err != nil {
-		// 	log.Log.Error(err, "Unable to patch ManagedJob")
-		// }
-		return true, err
-	}
-	return false, nil
-}
-
 func (cp *connPackage) updateCRDStatusDirectly() error {
 	cp.mtx.Lock()
-	// defer cp.mtx.Unlock()
-
-	// val, err := cp.getLatestMainJobAndPatch(cp.mj)
-	// if err != nil {
-	// 	log.Log.Error(err, "Unable to get latest ManagedJob")
-	// 	return err
-	// }
-
-	// if !val && err == nil {
 	err := cp.r.Update(cp.ctx, cp.mj)
 	if err != nil {
 		// log.Log.Info("Error", err.Error(), "more", "Unable to update ManagedJob status directly")
@@ -121,7 +59,6 @@ func (cp *connPackage) updateCRDStatusDirectly() error {
 	if err != nil {
 		log.Log.Error(err, "Unable to get updated ManagedJob")
 	}
-	// }
 	cp.mtx.Unlock()
 	return err
 }

controllers/managedjob_controller.go

@@ -19,6 +19,7 @@ package controllers
 import (
 	"context"
 
+	"github.com/lukaszraczylo/pandati"
 	kbatch "k8s.io/api/batch/v1"
 	"k8s.io/apimachinery/pkg/runtime"
 	"k8s.io/client-go/tools/record"
@@ -44,9 +45,10 @@ func (r *ManagedJobReconciler) Reconcile(ctx context.Context, req ctrl.Request)
 	_ = log.FromContext(ctx)
 
 	cp := &connPackage{
-		r:   r,
-		ctx: ctx,
-		req: req,
+		r:              r,
+		ctx:            ctx,
+		req:            req,
+		dependencyTree: nil,
 	}
 
 	var managedJob jobsmanagerv1beta1.ManagedJob
@@ -55,13 +57,26 @@ func (r *ManagedJobReconciler) Reconcile(ctx context.Context, req ctrl.Request)
 	}
 
 	cp.mj = &managedJob
-	cp.buildJobsDependencyTree()
-	cp.checkJobStatus()
+
+	originalMainJobDefinition := cp.mj.DeepCopy()
+	cp.generateDependencyTree()
+	_, theSame, _ := pandati.CompareStructsReplaced(originalMainJobDefinition, cp.mj)
+	if !theSame {
+		cp.updateCRDStatusDirectly()
+	}
+	originalMainJobDefinition = cp.mj.DeepCopy()
+
+	// TODO: Re-enable after testing
+	cp.checkRunningJobsStatus()
 	cp.runPendingJobs()
-	cp.checkGroupsStatus()
 
+	_, theSame, _ = pandati.CompareStructsReplaced(originalMainJobDefinition, cp.mj)
+	if !theSame {
+		cp.updateCRDStatusDirectly()
+	}
+
+	cp.checkOverallStatus()
 	// fmt.Printf("Reconcile: %# v", pretty.Formatter(r.Updater))
-
 	return ctrl.Result{}, nil
 }
 

go.mod

@@ -47,7 +47,7 @@ require (
 	github.com/pkg/errors v0.9.1 // indirect
 	github.com/prometheus/client_golang v1.14.0 // indirect
 	github.com/prometheus/client_model v0.3.0 // indirect
-	github.com/prometheus/common v0.39.0 // indirect
+	github.com/prometheus/common v0.40.0 // indirect
 	github.com/prometheus/procfs v0.9.0 // indirect
 	github.com/rogpeppe/go-internal v1.9.0 // indirect
 	github.com/rs/zerolog v1.29.0 // indirect
@@ -75,7 +75,7 @@ require (
 	k8s.io/component-base v0.26.1 // indirect
 	k8s.io/klog/v2 v2.90.0 // indirect
 	k8s.io/kube-openapi v0.0.0-20230217203603-ff9a8e8fa21d // indirect
-	k8s.io/utils v0.0.0-20230209194617-a36077c30491 // indirect
+	k8s.io/utils v0.0.0-20230220204549-a5ecb0141aa5 // indirect
 	sigs.k8s.io/json v0.0.0-20221116044647-bc3834ca7abd // indirect
 	sigs.k8s.io/structured-merge-diff/v4 v4.2.3 // indirect
 	sigs.k8s.io/yaml v1.3.0 // indirect

go.sum

@@ -292,6 +292,8 @@ github.com/prometheus/common v0.37.0 h1:ccBbHCgIiT9uSoFY0vX8H3zsNR5eLt17/RQLUvn8
 github.com/prometheus/common v0.37.0/go.mod h1:phzohg0JFMnBEFGxTDbfu3QyL5GI8gTQJFhYO5B3mfA=
 github.com/prometheus/common v0.39.0 h1:oOyhkDq05hPZKItWVBkJ6g6AtGxi+fy7F4JvUV8uhsI=
 github.com/prometheus/common v0.39.0/go.mod h1:6XBZ7lYdLCbkAVhwRsWTZn+IN5AB9F/NXd5w0BbEX0Y=
+github.com/prometheus/common v0.40.0 h1:Afz7EVRqGg2Mqqf4JuF9vdvp1pi220m55Pi9T2JnO4Q=
+github.com/prometheus/common v0.40.0/go.mod h1:L65ZJPSmfn/UBWLQIHV7dBrKFidB/wPlF1y5TlSt9OE=
 github.com/prometheus/procfs v0.0.0-20181005140218-185b4288413d/go.mod h1:c3At6R/oaqEKCNdg8wHV1ftS6bRYblBhIjjI8uT2IGk=
 github.com/prometheus/procfs v0.0.2/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsTZCD3I8kEA=
 github.com/prometheus/procfs v0.1.3/go.mod h1:lV6e/gmhEcM9IjHGsFOCxxuZ+z1YqCvr4OA4YeYWdaU=
@@ -725,6 +727,8 @@ k8s.io/utils v0.0.0-20221128185143-99ec85e7a448 h1:KTgPnR10d5zhztWptI952TNtt/4u5
 k8s.io/utils v0.0.0-20221128185143-99ec85e7a448/go.mod h1:OLgZIPagt7ERELqWJFomSt595RzquPNLL48iOWgYOg0=
 k8s.io/utils v0.0.0-20230209194617-a36077c30491 h1:r0BAOLElQnnFhE/ApUsg3iHdVYYPBjNSSOMowRZxxsY=
 k8s.io/utils v0.0.0-20230209194617-a36077c30491/go.mod h1:OLgZIPagt7ERELqWJFomSt595RzquPNLL48iOWgYOg0=
+k8s.io/utils v0.0.0-20230220204549-a5ecb0141aa5 h1:kmDqav+P+/5e1i9tFfHq1qcF3sOrDp+YEkVDAHu7Jwk=
+k8s.io/utils v0.0.0-20230220204549-a5ecb0141aa5/go.mod h1:OLgZIPagt7ERELqWJFomSt595RzquPNLL48iOWgYOg0=
 rsc.io/binaryregexp v0.2.0/go.mod h1:qTv7/COck+e2FymRvadv62gMdZztPaShugOCi3I+8D8=
 rsc.io/quote/v3 v3.1.0/go.mod h1:yEA65RcK8LyAZtP9Kv3t0HmxON59tX3rD+tICJqUlj0=
 rsc.io/sampler v1.3.0/go.mod h1:T1hPZKmBbMNahiBKFy5HrXp6adAjACjK9JXDnKaTXpA=