Support JobBackoffLimitPerIndex feature gate fields (#2421)

* update job spec

* fix tests by adding in extra checks and DiffSuppressFuncs for both new attributes

* add changelog

* add docs

* add new fields into TestAccKubernetesCronJobV1_minimalWithPodFailurePolicy test

* insert field checks for new fields

* make tests

* add flatteners

* lint fix

* add new test - TestAccKubernetesCronJobV1_minimalWithBackoffLimitPerIndex

* remove fields from other test

---------

Co-authored-by: Mauricio Alvarez Leon <[email protected]>
Co-authored-by: BBBmau <[email protected]>

Author: theloneexplorerquest

.changelog/2421.txt

@@ -0,0 +1,3 @@
+```release-note:enhancement
+Add `backoff_per_limit_index` and `max_failed_indexes` fields in `structure_job.go`
+```

docs/resources/cron_job.md

@@ -102,6 +102,8 @@ Optional:
 
 - `active_deadline_seconds` (Number) Optional duration in seconds the pod may be active on the node relative to StartTime before the system will actively try to mark it failed and kill associated containers. Value must be a positive integer.
 - `backoff_limit` (Number) Specifies the number of retries before marking this job failed. Defaults to 6
+- `backoff_limit_per_index` - (Number) Specifies the limit for the number of retries within an index before marking this index as failed. When enabled the number of failures per index is kept in the pod's batch.kubernetes.io/job-index-failure-count annotation. It can only be set when Job's completionMode=Indexed, and the Pod's restart policy is Never. The field is immutable.
+- `max_failed_indexes` - (Number) Controls generation of pod labels and pod selectors. Leave unset unless you are certain what you are doing. When false or unset, the system pick labels unique to this job and appends those labels to the pod template. When true, the user is responsible for picking unique labels and specifying the selector. Failure to pick a unique label may cause this and other jobs to not function correctly. More info: https://git.k8s.io/community/contributors/design-proposals/selector-generation.md
 - `completion_mode` (String) Specifies how Pod completions are tracked. It can be `NonIndexed` (default) or `Indexed`. More info: https://kubernetes.io/docs/concepts/workloads/controllers/job/#completion-mode
 - `completions` (Number) Specifies the desired number of successfully finished pods the job should be run with. Setting to nil means that the success of any pod signals the success of all pods, and allows parallelism to have any positive value. Setting to 1 means that parallelism is limited to 1 and the success of that pod signals the success of the job. More info: https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/
 - `manual_selector` (Boolean) Controls generation of pod labels and pod selectors. Leave unset unless you are certain what you are doing. When false or unset, the system pick labels unique to this job and appends those labels to the pod template. When true, the user is responsible for picking unique labels and specifying the selector. Failure to pick a unique label may cause this and other jobs to not function correctly. More info: https://git.k8s.io/community/contributors/design-proposals/selector-generation.md

docs/resources/cron_job_v1.md

@@ -97,6 +97,8 @@ Optional:
 
 - `active_deadline_seconds` (Number) Optional duration in seconds the pod may be active on the node relative to StartTime before the system will actively try to mark it failed and kill associated containers. Value must be a positive integer.
 - `backoff_limit` (Number) Specifies the number of retries before marking this job failed. Defaults to 6
+- `backoff_limit_per_index` - (Number) Specifies the limit for the number of retries within an index before marking this index as failed. When enabled the number of failures per index is kept in the pod's batch.kubernetes.io/job-index-failure-count annotation. It can only be set when Job's completionMode=Indexed, and the Pod's restart policy is Never. The field is immutable.
+- `max_failed_indexes` - (Number) Controls generation of pod labels and pod selectors. Leave unset unless you are certain what you are doing. When false or unset, the system pick labels unique to this job and appends those labels to the pod template. When true, the user is responsible for picking unique labels and specifying the selector. Failure to pick a unique label may cause this and other jobs to not function correctly. More info: https://git.k8s.io/community/contributors/design-proposals/selector-generation.md
 - `completion_mode` (String) Specifies how Pod completions are tracked. It can be `NonIndexed` (default) or `Indexed`. More info: https://kubernetes.io/docs/concepts/workloads/controllers/job/#completion-mode
 - `completions` (Number) Specifies the desired number of successfully finished pods the job should be run with. Setting to nil means that the success of any pod signals the success of all pods, and allows parallelism to have any positive value. Setting to 1 means that parallelism is limited to 1 and the success of that pod signals the success of the job. More info: https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/
 - `manual_selector` (Boolean) Controls generation of pod labels and pod selectors. Leave unset unless you are certain what you are doing. When false or unset, the system pick labels unique to this job and appends those labels to the pod template. When true, the user is responsible for picking unique labels and specifying the selector. Failure to pick a unique label may cause this and other jobs to not function correctly. More info: https://git.k8s.io/community/contributors/design-proposals/selector-generation.md

docs/resources/job_v1.md

@@ -55,6 +55,8 @@ Optional:
 
 - `active_deadline_seconds` (Number) Optional duration in seconds the pod may be active on the node relative to StartTime before the system will actively try to mark it failed and kill associated containers. Value must be a positive integer.
 - `backoff_limit` (Number) Specifies the number of retries before marking this job failed. Defaults to 6
+- `backoff_limit_per_index` - (Number) Specifies the limit for the number of retries within an index before marking this index as failed. When enabled the number of failures per index is kept in the pod's batch.kubernetes.io/job-index-failure-count annotation. It can only be set when Job's completionMode=Indexed, and the Pod's restart policy is Never. The field is immutable.
+- `max_failed_indexes` - (Number) Controls generation of pod labels and pod selectors. Leave unset unless you are certain what you are doing. When false or unset, the system pick labels unique to this job and appends those labels to the pod template. When true, the user is responsible for picking unique labels and specifying the selector. Failure to pick a unique label may cause this and other jobs to not function correctly. More info: https://git.k8s.io/community/contributors/design-proposals/selector-generation.md
 - `completion_mode` (String) Specifies how Pod completions are tracked. It can be `NonIndexed` (default) or `Indexed`. More info: https://kubernetes.io/docs/concepts/workloads/controllers/job/#completion-mode
 - `completions` (Number) Specifies the desired number of successfully finished pods the job should be run with. Setting to nil means that the success of any pod signals the success of all pods, and allows parallelism to have any positive value. Setting to 1 means that parallelism is limited to 1 and the success of that pod signals the success of the job. More info: https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/
 - `manual_selector` (Boolean) Controls generation of pod labels and pod selectors. Leave unset unless you are certain what you are doing. When false or unset, the system pick labels unique to this job and appends those labels to the pod template. When true, the user is responsible for picking unique labels and specifying the selector. Failure to pick a unique label may cause this and other jobs to not function correctly. More info: https://git.k8s.io/community/contributors/design-proposals/selector-generation.md

kubernetes/resource_kubernetes_cron_job_v1_test.go

@@ -222,6 +222,61 @@ func TestAccKubernetesCronJobV1_minimalWithPodFailurePolicy(t *testing.T) {
 	})
 }
 
+func TestAccKubernetesCronJobV1_minimalWithBackoffLimitPerIndex(t *testing.T) {
+	var conf1, conf2 batchv1.CronJob
+
+	name := fmt.Sprintf("tf-acc-test-%s", acctest.RandStringFromCharSet(10, acctest.CharSetAlphaNum))
+	resourceName := "kubernetes_cron_job_v1.test"
+	imageName := busyboxImage
+
+	resource.ParallelTest(t, resource.TestCase{
+		PreCheck: func() {
+			testAccPreCheck(t)
+			skipIfClusterVersionLessThan(t, "1.29.0")
+		},
+		IDRefreshName:     resourceName,
+		IDRefreshIgnore:   []string{"metadata.0.resource_version"},
+		ProviderFactories: testAccProviderFactories,
+		CheckDestroy:      testAccCheckKubernetesCronJobV1Destroy,
+		Steps: []resource.TestStep{
+			{
+				Config: testAccKubernetesCronJobV1ConfigMinimal(name, imageName),
+				Check: resource.ComposeAggregateTestCheckFunc(
+					testAccCheckKubernetesCronJobV1Exists(resourceName, &conf1),
+					resource.TestCheckResourceAttrSet(resourceName, "metadata.0.generation"),
+					resource.TestCheckResourceAttrSet(resourceName, "metadata.0.resource_version"),
+					resource.TestCheckResourceAttrSet(resourceName, "metadata.0.uid"),
+					resource.TestCheckResourceAttrSet(resourceName, "metadata.0.namespace"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.metadata.0.namespace", ""),
+				),
+			},
+			{
+				Config: testAccKubernetesCronJobV1ConfigMinimalWithBackoffLimitPerIndex(name, imageName),
+				Check: resource.ComposeAggregateTestCheckFunc(
+					testAccCheckKubernetesCronJobV1Exists(resourceName, &conf2),
+					resource.TestCheckResourceAttrSet(resourceName, "metadata.0.generation"),
+					resource.TestCheckResourceAttrSet(resourceName, "metadata.0.resource_version"),
+					resource.TestCheckResourceAttrSet(resourceName, "metadata.0.uid"),
+					resource.TestCheckResourceAttrSet(resourceName, "metadata.0.namespace"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.backoff_limit_per_index", "3"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.max_failed_indexes", "4"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.#", "2"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.0.action", "FailJob"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.container_name", "test"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.#", "3"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.0", "1"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.1", "2"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.2", "42"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.1.action", "Ignore"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.1.on_pod_condition.0.type", "DisruptionTarget"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.pod_failure_policy.0.rule.1.on_pod_condition.0.status", "False"),
+					testAccCheckKubernetesCronJobV1ForceNew(&conf1, &conf2, true),
+				),
+			},
+		},
+	})
+}
+
 func testAccCheckKubernetesCronJobV1Destroy(s *terraform.State) error {
 	conn, err := testAccProvider.Meta().(KubeClientsets).MainClientset()
 
@@ -461,6 +516,56 @@ func testAccKubernetesCronJobV1ConfigMinimal(name, imageName string) string {
 `, name, imageName)
 }
 
+func testAccKubernetesCronJobV1ConfigMinimalWithBackoffLimitPerIndex(name, imageName string) string {
+	return fmt.Sprintf(`resource "kubernetes_cron_job_v1" "test" {
+  metadata {
+    name = "%s"
+  }
+  spec {
+    schedule = "*/1 * * * *"
+    job_template {
+      metadata {}
+      spec {
+        backoff_limit_per_index = 3
+        max_failed_indexes      = 4
+        completions             = 4
+        completion_mode         = "Indexed"
+        pod_failure_policy {
+          rule {
+            action = "FailJob"
+            on_exit_codes {
+              container_name = "test"
+              operator       = "In"
+              values         = [1, 2, 42]
+            }
+          }
+          rule {
+            action = "Ignore"
+            on_pod_condition {
+              status = "False"
+              type   = "DisruptionTarget"
+            }
+          }
+        }
+        template {
+          metadata {}
+          spec {
+
+            container {
+              name    = "test"
+              image   = "%s"
+              command = ["sleep", "5"]
+            }
+            termination_grace_period_seconds = 1
+          }
+        }
+      }
+    }
+  }
+}
+`, name, imageName)
+}
+
 func testAccKubernetesCronJobV1ConfigMinimalWithPodFailurePolicy(name, imageName string) string {
 	return fmt.Sprintf(`resource "kubernetes_cron_job_v1" "test" {
   metadata {

kubernetes/schema_job_spec.go

@@ -57,6 +57,13 @@ func jobSpecFields(specUpdatable bool) map[string]*schema.Schema {
 			ValidateFunc: validateNonNegativeInteger,
 			Description:  "Specifies the number of retries before marking this job failed. Defaults to 6",
 		},
+		"backoff_limit_per_index": {
+			Type:         schema.TypeInt,
+			Optional:     true,
+			ForceNew:     true,
+			ValidateFunc: validateNonNegativeInteger,
+			Description:  "Specifies the limit for the number of retries within an index before marking this index as failed. When enabled the number of failures per index is kept in the pod's batch.kubernetes.io/job-index-failure-count annotation. It can only be set when Job's completionMode=Indexed, and the Pod's restart policy is Never. The field is immutable.",
+		},
 		// This field is immutable in Jobs.
 		"completions": {
 			Type:         schema.TypeInt,
@@ -83,6 +90,13 @@ func jobSpecFields(specUpdatable bool) map[string]*schema.Schema {
 			ForceNew:    false,
 			Description: "Controls generation of pod labels and pod selectors. Leave unset unless you are certain what you are doing. When false or unset, the system pick labels unique to this job and appends those labels to the pod template. When true, the user is responsible for picking unique labels and specifying the selector. Failure to pick a unique label may cause this and other jobs to not function correctly. More info: https://git.k8s.io/community/contributors/design-proposals/selector-generation.md",
 		},
+		"max_failed_indexes": {
+			Type:         schema.TypeInt,
+			Optional:     true,
+			ForceNew:     false,
+			ValidateFunc: validateNonNegativeInteger,
+			Description:  "Controls generation of pod labels and pod selectors. Leave unset unless you are certain what you are doing. When false or unset, the system pick labels unique to this job and appends those labels to the pod template. When true, the user is responsible for picking unique labels and specifying the selector. Failure to pick a unique label may cause this and other jobs to not function correctly. More info: https://git.k8s.io/community/contributors/design-proposals/selector-generation.md",
+		},
 		"parallelism": {
 			Type:         schema.TypeInt,
 			Optional:     true,

kubernetes/structure_job.go

@@ -23,6 +23,14 @@ func flattenJobV1Spec(in batchv1.JobSpec, d *schema.ResourceData, meta interface
 		att["backoff_limit"] = *in.BackoffLimit
 	}
 
+	if in.BackoffLimitPerIndex != nil {
+		att["backoff_limit_per_index"] = *in.BackoffLimitPerIndex
+	}
+
+	if in.MaxFailedIndexes != nil {
+		att["max_failed_indexes"] = *in.MaxFailedIndexes
+	}
+
 	if in.Completions != nil {
 		att["completions"] = *in.Completions
 	}
@@ -79,6 +87,10 @@ func expandJobV1Spec(j []interface{}) (batchv1.JobSpec, error) {
 		obj.BackoffLimit = ptr.To(int32(v))
 	}
 
+	if v, ok := in["backoff_limit_per_index"].(int); in["completion_mode"] == "Indexed" && ok && v >= 0 {
+		obj.BackoffLimitPerIndex = ptr.To(int32(v))
+	}
+
 	if v, ok := in["completions"].(int); ok && v > 0 {
 		obj.Completions = ptr.To(int32(v))
 	}
@@ -92,6 +104,10 @@ func expandJobV1Spec(j []interface{}) (batchv1.JobSpec, error) {
 		obj.ManualSelector = ptr.To(v.(bool))
 	}
 
+	if v, ok := in["max_failed_indexes"].(int); in["completion_mode"] == "Indexed" && ok && v >= 0 {
+		obj.MaxFailedIndexes = ptr.To(int32(v))
+	}
+
 	if v, ok := in["parallelism"].(int); ok && v >= 0 {
 		obj.Parallelism = ptr.To(int32(v))
 	}