diff --git a/pkg/health/health.go b/pkg/health/health.go
index dab6de6..e05097b 100644
--- a/pkg/health/health.go
+++ b/pkg/health/health.go
@@ -91,7 +91,6 @@ const (
 	HealthStatusEvicted          HealthStatusCode = "Evicted"
 	HealthStatusCompleted        HealthStatusCode = "Completed"
 	HealthStatusCrashLoopBackoff HealthStatusCode = "CrashLoopBackOff"
-	HealthStatusCrashLoop        HealthStatusCode = "CrashLoop"
 	HealthStatusCrashed          HealthStatusCode = "Crashed"
 	HealthStatusCreating         HealthStatusCode = "Creating"
 	HealthStatusDeleted          HealthStatusCode = "Deleted"
diff --git a/pkg/health/health_deployment.go b/pkg/health/health_deployment.go
index 0903804..69cae8f 100644
--- a/pkg/health/health_deployment.go
+++ b/pkg/health/health_deployment.go
@@ -80,7 +80,7 @@ func getReplicaHealth(s ReplicaStatus) *HealthStatus {
 		hs.Status = HealthStatusStarting
 	} else if s.Ready == 0 && !isStarting {
 		hs.Health = HealthUnhealthy
-		hs.Status = HealthStatusCrashLoop
+		hs.Status = HealthStatusCrashLoopBackoff
 	} else if s.Desired == 0 && s.Replicas > 0 {
 		hs.Status = HealthStatusScalingDown
 		hs.Health = lo.Ternary(isProgressDeadlineExceeded, HealthWarning, HealthHealthy)
diff --git a/pkg/health/health_pod.go b/pkg/health/health_pod.go
index 9197207..fe8bdc3 100644
--- a/pkg/health/health_pod.go
+++ b/pkg/health/health_pod.go
@@ -158,11 +158,16 @@ func getCorev1PodHealth(pod *corev1.Pod) (*HealthStatus, error) {
 		hr.Health = hr.Health.Worst(lo.Ternary(isReady, HealthHealthy, HealthUnhealthy))
 	}
 
-	if isStarting && hr.Health.IsWorseThan(HealthWarning) &&
-		(terminated != nil && terminated.Status != HealthStatusOOMKilled) {
+	if isStarting && hr.Health.IsWorseThan(HealthWarning) {
 		hr.Health = HealthUnknown
-		hr.Message = fmt.Sprintf("%s %s", string(hr.Status), hr.Message)
+		hr.Message = strings.TrimSpace(fmt.Sprintf("%s %s", string(hr.Status), hr.Message))
 		hr.Status = HealthStatusStarting
+
+		if terminated != nil && terminated.Status == HealthStatusOOMKilled {
+			// an OOMKilled on startup is likely not going to resolve after some time
+			hr.Health = HealthUnhealthy
+			hr.Status = "Starting OOMKilled"
+		}
 	}
 
 	return &hr, nil
diff --git a/pkg/health/health_test.go b/pkg/health/health_test.go
index 1139539..e17b26a 100644
--- a/pkg/health/health_test.go
+++ b/pkg/health/health_test.go
@@ -409,7 +409,7 @@ func TestStatefulSetHealth(t *testing.T) {
 	assertAppHealthMsg(
 		t,
 		"./testdata/statefulset-starting.yaml",
-		health.HealthStatusCrashLoop,
+		health.HealthStatusCrashLoopBackoff,
 		health.HealthUnhealthy,
 		true,
 		"0/1 ready",
@@ -419,7 +419,7 @@ func TestStatefulSetHealth(t *testing.T) {
 	assertAppHealthMsg(
 		t,
 		"./testdata/statefulset-starting.yaml",
-		health.HealthStatusCrashLoop,
+		health.HealthStatusCrashLoopBackoff,
 		health.HealthUnhealthy,
 		true,
 		"0/1 ready",
diff --git a/pkg/health/testdata/Kubernetes/Pod/container-creating.yaml b/pkg/health/testdata/Kubernetes/Pod/container-creating.yaml
new file mode 100644
index 0000000..b3dd84d
--- /dev/null
+++ b/pkg/health/testdata/Kubernetes/Pod/container-creating.yaml
@@ -0,0 +1,78 @@
+apiVersion: v1
+kind: Pod
+metadata:
+  creationTimestamp: "@now-5m"
+  name: image-pull-backoff
+  namespace: argocd
+  annotations:
+    expected-status: Starting
+    expected-health: unknown
+  resourceVersion: "155333"
+  selfLink: /api/v1/namespaces/argocd/pods/image-pull-backoff
+  uid: 46c1e8de-f61b-11e8-a057-fe5f49266390
+spec:
+  containers:
+  - image: does-not-exist
+    imagePullPolicy: Always
+    name: main
+    resources: {}
+    terminationMessagePath: /dev/termination-log
+    terminationMessagePolicy: File
+    volumeMounts:
+    - mountPath: /var/run/secrets/kubernetes.io/serviceaccount
+      name: default-token-f9jvj
+      readOnly: true
+  dnsPolicy: ClusterFirst
+  nodeName: minikube
+  restartPolicy: Always
+  schedulerName: default-scheduler
+  securityContext: {}
+  serviceAccount: default
+  serviceAccountName: default
+  terminationGracePeriodSeconds: 30
+  tolerations:
+  - effect: NoExecute
+    key: node.kubernetes.io/not-ready
+    operator: Exists
+    tolerationSeconds: 300
+  - effect: NoExecute
+    key: node.kubernetes.io/unreachable
+    operator: Exists
+    tolerationSeconds: 300
+  volumes:
+  - name: default-token-f9jvj
+    secret:
+      defaultMode: 420
+      secretName: default-token-f9jvj
+status:
+  conditions:
+  - lastProbeTime: null
+    lastTransitionTime: 2018-12-02T10:16:04Z
+    status: "True"
+    type: Initialized
+  - lastProbeTime: null
+    lastTransitionTime: 2018-12-02T10:16:04Z
+    message: 'containers with unready status: [main]'
+    reason: ContainersNotReady
+    status: "False"
+    type: Ready
+  - lastProbeTime: null
+    lastTransitionTime: 2018-12-02T10:16:04Z
+    status: "True"
+    type: PodScheduled
+  containerStatuses:
+  - image: does-not-exist
+    imageID: ""
+    lastState: {}
+    name: main
+    ready: false
+    started: false
+    restartCount: 0
+    state:
+      waiting:
+        reason: ContainerCreating
+  hostIP: 192.168.64.41
+  phase: Pending
+  podIP: 172.17.0.9
+  qosClass: BestEffort
+  startTime: 2018-12-02T10:16:04Z
diff --git a/pkg/health/testdata/Kubernetes/Pod/pod-pending.yaml b/pkg/health/testdata/Kubernetes/Pod/pod-pending.yaml
index 6a99ec1..151be39 100644
--- a/pkg/health/testdata/Kubernetes/Pod/pod-pending.yaml
+++ b/pkg/health/testdata/Kubernetes/Pod/pod-pending.yaml
@@ -66,6 +66,7 @@ status:
     lastState: {}
     name: main
     ready: false
+    started: true
     restartCount: 0
     state:
       waiting: