Add PodFailurePolicy for Jobs

.changelog/2394.txt

@@ -0,0 +1,3 @@
+```release-note:enhancement
+`resource/kubernetes_job_v1`: add new attribute `spec.pod_failure_policy` to job spec
+```

kubernetes/resource_kubernetes_cron_job_v1_test.go

@@ -57,6 +57,16 @@ func TestAccKubernetesCronJobV1_basic(t *testing.T) {
 					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.template.0.metadata.0.annotations.controller.kubernetes.io/pod-deletion-cost", "10000"),
 					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.template.0.spec.0.container.0.name", "hello"),
 					resource.TestCheckResourceAttr(resourceName, "spec.0.job_template.0.spec.0.template.0.spec.0.container.0.image", imageName),
+					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", "hello"),
+					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"),
 				),
 			},
 			{
@@ -255,6 +265,23 @@ func testAccKubernetesCronJobV1Config_basic(name, imageName string) string {
       }
       spec {
         backoff_limit = 2
+        pod_failure_policy {
+          rule {
+            action = "FailJob"
+            on_exit_codes {
+              container_name = "hello"
+              operator       = "In"
+              values         = [2, 1, 42]
+            }
+          }
+          rule {
+            action = "Ignore"
+            on_pod_condition {
+              status = "False"
+              type   = "DisruptionTarget"
+            }
+          }
+        }
         template {
           metadata {
             annotations = {

kubernetes/resource_kubernetes_job_v1_test.go

@@ -77,6 +77,16 @@ func TestAccKubernetesJobV1_basic(t *testing.T) {
 					resource.TestCheckResourceAttr(resourceName, "spec.0.template.0.spec.0.container.0.name", "hello"),
 					resource.TestCheckResourceAttr(resourceName, "spec.0.template.0.spec.0.container.0.image", imageName),
 					resource.TestCheckResourceAttr(resourceName, "wait_for_completion", "false"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.#", "2"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.action", "FailJob"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.container_name", "hello"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.#", "3"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.0", "1"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.1", "2"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.2", "42"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.1.action", "Ignore"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.1.on_pod_condition.0.type", "DisruptionTarget"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.1.on_pod_condition.0.status", "False"),
 				),
 			},
 			{
@@ -131,6 +141,13 @@ func TestAccKubernetesJobV1_update(t *testing.T) {
 					resource.TestCheckResourceAttr(resourceName, "spec.0.template.0.spec.0.container.0.image", imageName),
 					resource.TestCheckResourceAttr(resourceName, "wait_for_completion", "false"),
 					resource.TestCheckResourceAttr(resourceName, "spec.0.manual_selector", "false"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.#", "2"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.action", "FailJob"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.container_name", "hello"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.0.on_exit_codes.0.values.#", "3"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.1.action", "Ignore"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.1.on_pod_condition.0.type", "DisruptionTarget"),
+					resource.TestCheckResourceAttr(resourceName, "spec.0.pod_failure_policy.0.rule.1.on_pod_condition.0.status", "False"),
 				),
 			},
 			{
@@ -307,6 +324,23 @@ func testAccKubernetesJobV1Config_basic(name, imageName string) string {
     backoff_limit           = 10
     completions             = 4
     parallelism             = 2
+    pod_failure_policy {
+      rule {
+        action = "FailJob"
+        on_exit_codes {
+          container_name = "hello"
+          operator       = "In"
+          values         = [2, 1, 42]
+        }
+      }
+      rule {
+        action = "Ignore"
+        on_pod_condition {
+          status = "False"
+          type   = "DisruptionTarget"
+        }
+      }
+    }
     template {
       metadata {}
       spec {
@@ -334,6 +368,23 @@ func testAccKubernetesJobV1Config_updateMutableFields(name, imageName, activeDea
     completions             = 4
     manual_selector         = %s
     parallelism             = %s
+    pod_failure_policy {
+      rule {
+        action = "FailJob"
+        on_exit_codes {
+          container_name = "hello"
+          operator       = "In"
+          values         = [2, 1, 42]
+        }
+      }
+      rule {
+        action = "Ignore"
+        on_pod_condition {
+          status = "False"
+          type   = "DisruptionTarget"
+        }
+      }
+    }
     template {
       metadata {}
       spec {

kubernetes/schema_job_spec.go

@@ -91,6 +91,72 @@ func jobSpecFields(specUpdatable bool) map[string]*schema.Schema {
 			ValidateFunc: validateNonNegativeInteger,
 			Description:  "Specifies the maximum desired number of pods the job should run at any given time. The actual number of pods running in steady state will be less than this number when ((.spec.completions - .status.successful) < .spec.parallelism), i.e. when the work left to do is less than max parallelism. More info: https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/",
 		},
+		"pod_failure_policy": {
+			Type:        schema.TypeList,
+			Optional:    true,
+			ForceNew:    false,
+			MaxItems:    1,
+			Description: "Specifies the maximum desired number of pods the job should run at any given time. The actual number of pods running in steady state will be less than this number when ((.spec.completions - .status.successful) < .spec.parallelism), i.e. when the work left to do is less than max parallelism. More info: https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/",
+			Elem: &schema.Resource{
+				Schema: map[string]*schema.Schema{
+					"rule": {
+						Type:        schema.TypeList,
+						Description: "A label query over volumes to consider for binding.",
+						Required:    true,
+						Elem: &schema.Resource{
+							Schema: map[string]*schema.Schema{
+								"action": {
+									Type:     schema.TypeString,
+									Optional: true,
+								},
+								"on_exit_codes": {
+									Type:     schema.TypeList,
+									Optional: true,
+									MaxItems: 1,
+									Elem: &schema.Resource{
+										Schema: map[string]*schema.Schema{
+											"container_name": {
+												Type:     schema.TypeString,
+												Optional: true,
+											},
+											"operator": {
+												Type:     schema.TypeString,
+												Optional: true,
+											},
+											"values": {
+												Type:     schema.TypeList,
+												Required: true,
+												MinItems: 1,
+												MaxItems: 255,
+												Elem: &schema.Schema{Type: schema.TypeInt,
+													ValidateFunc: validation.IntNotInSlice([]int{0})},
+											},
+										},
+									},
+								},
+								"on_pod_condition": {
+									Type:     schema.TypeList,
+									Optional: true,
+									Elem: &schema.Resource{
+										Schema: map[string]*schema.Schema{
+											"status": {
+												Type:     schema.TypeString,
+												Optional: true,
+												Default:  "True",
+											},
+											"type": {
+												Type:     schema.TypeString,
+												Optional: true,
+											},
+										},
+									},
+								},
+							},
+						},
+					},
+				},
+			},
+		},
 		// This field is immutable in Jobs.
 		"selector": {
 			Type:        schema.TypeList,

kubernetes/structure_job.go

@@ -6,8 +6,11 @@ package kubernetes
 import (
 	"strconv"
 
+	"slices"
+
 	"github.com/hashicorp/terraform-plugin-sdk/v2/helper/schema"
 	batchv1 "k8s.io/api/batch/v1"
+	v1 "k8s.io/api/core/v1"
 	"k8s.io/utils/ptr"
 )
 
@@ -38,6 +41,10 @@ func flattenJobV1Spec(in batchv1.JobSpec, d *schema.ResourceData, meta interface
 		att["parallelism"] = *in.Parallelism
 	}
 
+	if in.PodFailurePolicy != nil {
+		att["pod_failure_policy"] = flattenPodFailurePolicy(in.PodFailurePolicy)
+	}
+
 	if in.Selector != nil {
 		att["selector"] = flattenLabelSelector(in.Selector)
 	}
@@ -91,6 +98,10 @@ func expandJobV1Spec(j []interface{}) (batchv1.JobSpec, error) {
 		obj.Parallelism = ptr.To(int32(v))
 	}
 
+	if v, ok := in["pod_failure_policy"].([]interface{}); ok && len(v) > 0 {
+		obj.PodFailurePolicy = expandPodFailurePolicy(v)
+	}
+
 	if v, ok := in["selector"].([]interface{}); ok && len(v) > 0 {
 		obj.Selector = expandLabelSelector(v)
 	}
@@ -112,6 +123,152 @@ func expandJobV1Spec(j []interface{}) (batchv1.JobSpec, error) {
 	return obj, nil
 }
 
+func expandPodFailurePolicy(l []interface{}) *batchv1.PodFailurePolicy {
+	obj := &batchv1.PodFailurePolicy{}
+	if len(l) == 0 || l[0] == nil {
+		return obj
+	}
+	in := l[0].(map[string]interface{})
+
+	if v, ok := in["rule"].([]interface{}); ok && len(v) > 0 {
+		rules := expandPodFailurePolicyRules(v)
+		obj.Rules = rules
+	}
+	return obj
+}
+
+func expandPodFailurePolicyRules(l []interface{}) []batchv1.PodFailurePolicyRule {
+	obj := make([]batchv1.PodFailurePolicyRule, len(l))
+	if len(l) == 0 || l[0] == nil {
+		return obj
+	}
+	for i, rule := range l {
+		objRule := &batchv1.PodFailurePolicyRule{}
+
+		r := rule.(map[string]interface{})
+
+		if v, ok := r["action"].(string); ok && v != "" {
+			objRule.Action = batchv1.PodFailurePolicyAction(v)
+		}
+
+		if v, ok := r["on_exit_codes"].([]interface{}); ok && len(v) > 0 {
+			onExitCodes := expandPodFailurePolicyOnExitCodesRequirement(v)
+			objRule.OnExitCodes = onExitCodes
+		}
+
+		if v, ok := r["on_pod_condition"].([]interface{}); ok && len(v) > 0 {
+			podConditions := expandPodFailurePolicyOnPodConditionsPattern(v)
+			objRule.OnPodConditions = podConditions
+		}
+
+		obj[i] = *objRule
+	}
+
+	return obj
+}
+
+func expandPodFailurePolicyOnExitCodesRequirement(l []interface{}) *batchv1.PodFailurePolicyOnExitCodesRequirement {
+	obj := &batchv1.PodFailurePolicyOnExitCodesRequirement{}
+	if len(l) == 0 || l[0] == nil {
+		return obj
+	}
+	in := l[0].(map[string]interface{})
+
+	if v, ok := in["container_name"].(string); ok && v != "" {
+		obj.ContainerName = &v
+	}
+
+	if v, ok := in["operator"].(string); ok && v != "" {
+		obj.Operator = batchv1.PodFailurePolicyOnExitCodesOperator(v)
+	}
+
+	if v, ok := in["values"].(*schema.Set); ok && v != nil {
+		v := schemaSetToInt32Array(v)
+		slices.Sort(v)
+		obj.Values = v
+	}
+
+	return obj
+}
+
+func expandPodFailurePolicyOnPodConditionsPattern(l []interface{}) []batchv1.PodFailurePolicyOnPodConditionsPattern {
+	obj := make([]batchv1.PodFailurePolicyOnPodConditionsPattern, len(l))
+	if len(l) == 0 || l[0] == nil {
+		return obj
+	}
+	for i, condition := range l {
+		objCondition := &batchv1.PodFailurePolicyOnPodConditionsPattern{}
+		c := condition.(map[string]interface{})
+		if v, ok := c["status"].(string); ok && v != "" {
+			objCondition.Status = v1.ConditionStatus(v)
+		}
+
+		if v, ok := c["type"].(string); ok && v != "" {
+			objCondition.Type = v1.PodConditionType(v)
+		}
+		obj[i] = *objCondition
+	}
+	return obj
+}
+
+func flattenPodFailurePolicy(in *batchv1.PodFailurePolicy) []interface{} {
+	att := make(map[string]interface{})
+	if len(in.Rules) > 0 {
+		att["rule"] = flattenPodFailurePolicyRules(in.Rules)
+	}
+	return []interface{}{att}
+}
+
+func flattenPodFailurePolicyRules(in []batchv1.PodFailurePolicyRule) []interface{} {
+	att := make([]interface{}, len(in))
+
+	for i, r := range in {
+		m := make(map[string]interface{})
+		m["action"] = r.Action
+		if r.OnExitCodes != nil {
+			m["on_exit_codes"] = flattenPodFailurePolicyOnExitCodes(r.OnExitCodes)
+		}
+		if r.OnPodConditions != nil {
+			m["on_pod_condition"] = flattenPodFailurePolicyOnPodConditions(r.OnPodConditions)
+		}
+		att[i] = m
+	}
+
+	return att
+}
+
+func flattenPodFailurePolicyOnExitCodes(in *batchv1.PodFailurePolicyOnExitCodesRequirement) []interface{} {
+	att := make(map[string]interface{})
+	if *in.ContainerName != "" {
+		att["container_name"] = *in.ContainerName
+	}
+	att["operator"] = in.Operator
+	if len(in.Values) > 0 {
+		att["values"] = newInt32Set(schema.HashSchema(&schema.Schema{
+			Type: schema.TypeInt,
+		}), in.Values)
+	}
+
+	return []interface{}{att}
+}
+
+func flattenPodFailurePolicyOnPodConditions(in []batchv1.PodFailurePolicyOnPodConditionsPattern) []interface{} {
+	att := make([]interface{}, len(in))
+
+	for i, r := range in {
+		m := make(map[string]interface{})
+		if r.Status != "" {
+			m["status"] = r.Status
+		}
+		if r.Type != "" {
+			m["type"] = r.Type
+		}
+		att[i] = m
+	}
+
+	return att
+}
+
 func patchJobV1Spec(pathPrefix, prefix string, d *schema.ResourceData) PatchOperations {
 	ops := make([]PatchOperation, 0)