feat: Add health check for HPA (#4915)

HorizontalPodAutoscaler uses annotations to store conditions. This commit adds a custom heath check that parses the conditions from annotation and creates health status.

Fixes: #4413

Signed-off-by: Chetan Banavikalmutt <chetanrns1997@gmail.com>

resource_customizations/autoscaling/HorizontalPodAutoscaler/health.lua

@@ -0,0 +1,47 @@
+local json = require("json")
+health_status = {
+    status = "Progressing",
+    message = "Waiting to Autoscale"
+}
+
+function is_degraded( condition )
+    degraded_states = {
+        {"AbleToScale", "FailedGetScale"},
+        {"AbleToScale", "FailedUpdateScale"},
+        {"ScalingActive", "FailedGetResourceMetric"},
+        {"ScalingActive", "InvalidSelector"}
+    }
+    for i, state in ipairs(degraded_states) do
+        if condition.type == state[1] and condition.reason == state[2] then
+            return true
+        end 
+    end
+    return false
+end
+
+condition_string = obj.metadata.annotations["autoscaling.alpha.kubernetes.io/conditions"]
+if condition_string ~= nil then
+    conditions = json.decode(condition_string)
+end
+
+if conditions then
+    for i, condition in ipairs(conditions) do
+        if is_degraded(condition) then
+            health_status.status = "Degraded"
+            health_status.message = condition.message
+            return health_status
+        end
+        if condition.type == "AbleToScale" and  condition.reason == "SucceededRescale" then
+            health_status.status = "Healthy"
+            health_status.message = condition.message
+            return health_status
+        end
+        if condition.type == "ScalingLimited" and  condition.reason == "DesiredWithinRange" then
+            health_status.status = "Healthy"
+            health_status.message = condition.message
+            return health_status
+        end
+    end
+end
+
+return health_status

resource_customizations/autoscaling/HorizontalPodAutoscaler/health_test.yaml

@@ -0,0 +1,13 @@
+tests:
+- healthStatus:
+    status: Progressing
+    message: Waiting to Autoscale
+  inputPath: testdata/progressing_hpa.yaml
+- healthStatus:
+    status: Healthy
+    message:  the HPA controller was able to update the target scale to 1
+  inputPath: testdata/healthy_hpa.yaml
+- healthStatus:
+    status: Degraded
+    message: the HPA controller was unable to get the target's current scale
+  inputPath: testdata/degraded_hpa.yaml

resource_customizations/autoscaling/HorizontalPodAutoscaler/testdata/degraded_hpa.yaml

@@ -0,0 +1,21 @@
+apiVersion: autoscaling/v1
+kind: HorizontalPodAutoscaler
+metadata:
+  annotations:
+    autoscaling.alpha.kubernetes.io/conditions: '[{"type":"AbleToScale","status":"True","lastTransitionTime":"2020-11-23T19:38:38Z","reason":"FailedGetScale","message":"the HPA controller was unable to get the target''s current scale"},{"type":"ScalingActive","status":"False","lastTransitionTime":"2020-11-23T19:38:38Z","reason":"FailedGetResourceMetric","message":"the
+      HPA was unable to compute the replica count: unable to get metrics for resource
+      cpu: unable to fetch metrics from resource metrics API: the server is currently
+      unable to handle the request (get pods.metrics.k8s.io)"}]'
+  name: sample
+  namespace: argocd
+spec:
+  maxReplicas: 1
+  minReplicas: 1
+  scaleTargetRef:
+    apiVersion: apps/v1
+    kind: Deployment
+    name: sample
+  targetCPUUtilizationPercentage: 2
+status:
+  currentReplicas: 1
+  desiredReplicas: 0

resource_customizations/autoscaling/HorizontalPodAutoscaler/testdata/healthy_hpa.yaml

@@ -0,0 +1,21 @@
+apiVersion: autoscaling/v1
+kind: HorizontalPodAutoscaler
+metadata:
+  annotations:
+    autoscaling.alpha.kubernetes.io/conditions: '[{"type":"AbleToScale","status":"True","lastTransitionTime":"2020-11-23T19:38:38Z","reason":"SucceededRescale","message":"the HPA controller was able to update the target scale to 1"},{"type":"ScalingActive","status":"False","lastTransitionTime":"2020-11-23T19:38:38Z","reason":"FailedGetResourceMetric","message":"the
+      HPA was unable to compute the replica count: unable to get metrics for resource
+      cpu: unable to fetch metrics from resource metrics API: the server is currently
+      unable to handle the request (get pods.metrics.k8s.io)"}]'
+  name: sample
+  namespace: argocd
+spec:
+  maxReplicas: 2
+  minReplicas: 1
+  scaleTargetRef:
+    apiVersion: apps/v1
+    kind: Deployment
+    name: sample
+  targetCPUUtilizationPercentage: 2
+status:
+  currentReplicas: 1
+  desiredReplicas: 1

resource_customizations/autoscaling/HorizontalPodAutoscaler/testdata/progressing_hpa.yaml

@@ -0,0 +1,18 @@
+apiVersion: autoscaling/v1
+kind: HorizontalPodAutoscaler
+metadata:
+  annotations:
+    autoscaling.alpha.kubernetes.io/conditions: '[{"type":"AbleToScale","status":"True","lastTransitionTime":"2020-11-23T19:38:38Z","reason":"SucceededGetScale","message":"the HPA controller was able to get the target''s current scale"}]'
+  name: sample
+  namespace: argocd
+spec:
+  maxReplicas: 1
+  minReplicas: 1
+  scaleTargetRef:
+    apiVersion: apps/v1
+    kind: Deployment
+    name: sample
+  targetCPUUtilizationPercentage: 2
+status:
+  currentReplicas: 1
+  desiredReplicas: 0

util/lua/lua.go

@@ -88,6 +88,9 @@ func (vm VM) runLua(obj *unstructured.Unstructured, script string) (*lua.LState,
 	// preload our 'safe' version of the os library. Allows the 'local os = require("os")' to work
 	l.PreloadModule(lua.OsLibName, SafeOsLoader)
 
+	// preload json library to parse json in lua
+	luajson.Preload(l)
+
 	ctx, cancel := context.WithTimeout(context.Background(), 1*time.Second)
 	defer cancel()
 	l.SetContext(ctx)