From 70595cd7e0754b98c7d2a8b6d1b8bc0b9d0134b8 Mon Sep 17 00:00:00 2001
From: Lukasz Raczylo <lukasz@raczylo.com>
Date: Fri, 20 Dec 2024 11:34:18 +0000
Subject: [PATCH] fixup! fixup! fixup! fixup! fixup! fixup! fixup! Ensure that
 jobs run with the controller service account if no service account is
 specified.

---
 chart/Chart.yaml                              |  4 ++--
 chart/values.yaml                             |  2 +-
 .../raczylo.com/clusterimage_controller.go    | 21 ++++++++++++++++---
 internal/shared/jobs.go                       |  1 +
 4 files changed, 22 insertions(+), 6 deletions(-)

diff --git a/chart/Chart.yaml b/chart/Chart.yaml
index a838274..46c7473 100644
--- a/chart/Chart.yaml
+++ b/chart/Chart.yaml
@@ -10,9 +10,9 @@ description: |
 
 type: application
 
-version: 0.2.22
+version: 0.2.24
 
-appVersion: "0.2.22"
+appVersion: "0.2.24"
 
 home: https://github.com/lukaszraczylo/kubernetes-images-sync-operator
 
diff --git a/chart/values.yaml b/chart/values.yaml
index 0666733..e66365d 100644
--- a/chart/values.yaml
+++ b/chart/values.yaml
@@ -12,7 +12,7 @@ sa:
         - ALL
     image:
       repository: ghcr.io/lukaszraczylo/kubernetes-images-sync-operator
-      tag: 0.2.22
+      tag: 0.2.24
     resources:
       limits:
         cpu: 500m
diff --git a/internal/controller/raczylo.com/clusterimage_controller.go b/internal/controller/raczylo.com/clusterimage_controller.go
index 21eb73c..c6cd023 100644
--- a/internal/controller/raczylo.com/clusterimage_controller.go
+++ b/internal/controller/raczylo.com/clusterimage_controller.go
@@ -138,12 +138,27 @@ func (r *ClusterImageReconciler) handleRunningClusterImage(ctx context.Context,
 
 	if err != nil {
 		if errors.IsNotFound(err) {
-			// Job doesn't exist, set status back to PENDING
-			clusterImage.Status.Progress = shared.STATUS_PENDING
+			l.Info("Job not found; assuming it completed successfully", "job", jobName)
+
+			if clusterImage.Status.Progress == shared.STATUS_SUCCESS || clusterImage.Status.Progress == shared.STATUS_PRESENT {
+				// Job completed and status is already SUCCESS or PRESENT
+				return ctrl.Result{}, nil
+			}
+
+			// If we have retries left, consider retrying
+			if clusterImage.Status.RetryCount < 3 {
+				clusterImage.Status.Progress = shared.STATUS_RETRYING
+				clusterImage.Status.RetryCount++
+			} else {
+				// Exceeded retries; mark as FAILED
+				clusterImage.Status.Progress = shared.STATUS_FAILED
+			}
+
 			if err := r.Status().Update(ctx, clusterImage); err != nil {
-				l.Error(err, "unable to update ClusterImage status back to PENDING")
+				l.Error(err, "unable to update ClusterImage status after job not found")
 				return ctrl.Result{}, err
 			}
+
 			return ctrl.Result{Requeue: true}, nil
 		}
 		l.Error(err, "unable to check for existing job")
diff --git a/internal/shared/jobs.go b/internal/shared/jobs.go
index cd57f34..58862ee 100644
--- a/internal/shared/jobs.go
+++ b/internal/shared/jobs.go
@@ -62,6 +62,7 @@ func CreateJob[T any](params JobParams, setupFunc func(T) []string) *batchv1.Job
 			Annotations: params.Annotations,
 		},
 		Spec: batchv1.JobSpec{
+			TTLSecondsAfterFinished: pointer.Int32(300),
 			Template: corev1.PodTemplateSpec{
 				ObjectMeta: metav1.ObjectMeta{
 					Labels: map[string]string{