Utilize the DruidIngestion controller in e2e tests

datainfrahq · Mar 2, 2024 · 3f355a0 · 3f355a0
1 parent 8b1f785
commit 3f355a0
Show file tree

Hide file tree

Showing 9 changed files with 173 additions and 5 deletions.
diff --git a/Makefile b/Makefile
@@ -92,7 +92,12 @@ docker-push-local-test: ## Push docker image with the manager to kind registry.
 .PHONY: deploy-testjob
 deploy-testjob: ## Run a wikipedia test pod
 	kubectl create job wiki-test --image=${IMG_KIND}:${TEST_IMG_TAG}  -- sh /wikipedia-test.sh
-	bash e2e/monitor-task.sh
+	JOB_ID="wiki-test" bash e2e/monitor-task.sh
+
+.PHONY: deploy-testingestionjob
+deploy-testingestionjob: ## wait for the druidIngestion to complete and then verify dataset
+	kubectl create job ingestion-test --image=${IMG_KIND}:${TEST_IMG_TAG}  -- sh /druid-ingestion-test.sh ${TASK_ID}
+	JOB_ID="ingestion-test" bash e2e/monitor-task.sh
 
 .PHONY: helm-install-druid-operator
 helm-install-druid-operator: ## Helm install to deploy the druid operator

diff --git a/chart/templates/rbac_manager.yaml b/chart/templates/rbac_manager.yaml
@@ -126,6 +126,26 @@ rules:
     - get
     - patch
     - update
+- apiGroups:
+    - druid.apache.org
+  resources:
+    - druidingestions
+  verbs:
+    - create
+    - delete
+    - get
+    - list
+    - patch
+    - update
+    - watch
+- apiGroups:
+    - druid.apache.org
+  resources:
+    - druidingestions/status
+  verbs:
+    - get
+    - patch
+    - update
 - apiGroups:
     - networking.k8s.io
   resources:
@@ -284,6 +304,26 @@ rules:
     - get
     - patch
     - update
+- apiGroups:
+    - druid.apache.org
+  resources:
+    - druidingestions
+  verbs:
+    - create
+    - delete
+    - get
+    - list
+    - patch
+    - update
+    - watch
+- apiGroups:
+    - druid.apache.org
+  resources:
+    - druidingestions/status
+  verbs:
+    - get
+    - patch
+    - update
 - apiGroups:
     - networking.k8s.io
   resources:

diff --git a/controllers/ingestion/reconciler.go b/controllers/ingestion/reconciler.go
@@ -345,8 +345,8 @@ func (r *DruidIngestionReconciler) getRouterSvcUrl(namespace, druidClusterName s
 	if svcName == "" {
 		return "", errors.New("router svc discovery fail")
 	}
-	//	newName := "http://" + svcName + "." + namespace + ".svc.cluster.local:" + DruidRouterPort
-	newName := "http://localhost:" + DruidRouterPort
+
+	newName := "http://" + svcName + "." + namespace + ".svc.cluster.local:" + DruidRouterPort
 
 	return newName, nil
 }

diff --git a/e2e/Dockerfile-testpod b/e2e/Dockerfile-testpod
@@ -4,3 +4,4 @@ RUN  apk add --update-cache \
     && rm -rf /var/cache/apk/*
 
 ADD e2e/wikipedia-test.sh .
+ADD e2e/druid-ingestion-test.sh .
diff --git a/e2e/configs/druid-ingestion-cr.yaml b/e2e/configs/druid-ingestion-cr.yaml
@@ -0,0 +1,73 @@
+apiVersion: druid.apache.org/v1alpha1
+kind: DruidIngestion
+metadata:
+  labels:
+    app.kubernetes.io/name: druidingestion
+    app.kubernetes.io/instance: druidingestion-sample
+  name: wikipedia-ingestion
+spec:
+  suspend: false
+  druidCluster: tiny-cluster
+  ingestion:
+    type: native-batch
+    spec: |-
+      {
+        "type" : "index_parallel",
+        "spec" : {
+          "dataSchema" : {
+            "dataSource" : "wikipedia-2",
+            "timestampSpec": {
+              "column": "time",
+              "format": "iso"
+            },
+            "dimensionsSpec" : {
+              "dimensions" : [
+                "channel",
+                "cityName",
+                "comment",
+                "countryIsoCode",
+                "countryName",
+                "isAnonymous",
+                "isMinor",
+                "isNew",
+                "isRobot",
+                "isUnpatrolled",
+                "metroCode",
+                "namespace",
+                "page",
+                "regionIsoCode",
+                "regionName",
+                "user",
+                { "name": "added", "type": "long" },
+                { "name": "deleted", "type": "long" },
+                { "name": "delta", "type": "long" }
+              ]
+            },
+            "metricsSpec" : [],
+            "granularitySpec" : {
+              "type" : "uniform",
+              "segmentGranularity" : "day",
+              "queryGranularity" : "none",
+              "intervals" : ["2015-09-12/2015-09-13"],
+              "rollup" : false
+            }
+          },
+          "ioConfig" : {
+            "type" : "index_parallel",
+            "inputSource" : {
+              "type" : "local",
+              "baseDir" : "quickstart/tutorial/",
+              "filter" : "wikiticker-2015-09-12-sampled.json.gz"
+            },
+            "inputFormat" : {
+              "type" : "json"
+            },
+            "appendToExisting" : false
+          },
+          "tuningConfig" : {
+            "type" : "index_parallel",
+            "maxRowsPerSegment" : 5000000,
+            "maxRowsInMemory" : 25000
+          }
+        }
+      }
diff --git a/e2e/druid-ingestion-test.sh b/e2e/druid-ingestion-test.sh
@@ -0,0 +1,40 @@
+#!/bin/sh
+
+set -e
+
+TASK_ID=$1
+
+echo "Checking Status for task $TASK_ID..."
+STATUS=$(curl -s http://druid-tiny-cluster-coordinators.druid.svc:8088/druid/indexer/v1/task/${TASK_ID}/status | jq '.status.status' -r); 
+while [ $STATUS == "RUNNING" ]
+do 
+    sleep 8;
+    echo "TASK is "$STATUS "..."
+    STATUS=$(curl -s http://druid-tiny-cluster-coordinators.druid.svc:8088/druid/indexer/v1/task/${TASK_ID}/status | jq '.status.status' -r)
+done
+
+if [ $STATUS == "SUCCESS" ]
+then 
+    echo "TASK $TASK_ID COMPLETED SUCCESSFULLY"
+    sleep 60 # need time for the segments to become queryable
+else 
+    echo "TASK $TASK_ID  FAILED !!!!"
+    exit 1
+fi
+
+echo "Querying Data ... "
+echo "Running query SELECT COUNT(*) AS \"Count\" FROM \"wikipedia-2\" WHERE isMinor = 'false'"
+
+cat > query.json <<EOF
+{"query":"SELECT COUNT(*) AS \"Count\" FROM \"wikipedia-2\" WHERE isMinor = 'false'","resultFormat":"objectlines"}
+EOF
+
+count=`curl -s -XPOST -H'Content-Type: application/json' http://druid-tiny-cluster-routers.druid.svc:8088/druid/v2/sql -d @query.json| jq '.Count'`
+echo "count is $count"
+if [ $count != "21936" ]
+then
+    echo "Query failed !!!"
+    exit 1
+else
+    echo "Query Successful !!!"
+fi
diff --git a/e2e/e2e.sh b/e2e/e2e.sh
@@ -56,6 +56,14 @@ done
 # Running test job with an example dataset
 make deploy-testjob
 
+# Running a test DruidIngestion resource and wait for the task to be submitted
+kubectl apply -f e2e/configs/druid-ingestion-cr.yaml -n ${NAMESPACE}
+sleep 30 # wait for the manager to submit the ingestion task
+
+# get the ingestion task ID and launch the monitoring job
+taskId=`kubectl get druidingestion -n druid wikipedia-ingestion --template={{.status.taskId}}`
+make deploy-testingestionjob TASK_ID=$taskId
+
 # Delete old druid
 kubectl delete -f e2e/configs/druid-cr.yaml -n ${NAMESPACE}
 for d in $(kubectl get pods -n ${NAMESPACE} -l app=druid -l druid_cr=tiny-cluster -o name)

diff --git a/e2e/monitor-task.sh b/e2e/monitor-task.sh
@@ -2,11 +2,11 @@
 #!/bin/sh
 set -e 
 echo "---------------"
-echo "Checking the status of running job ..."
+echo "Checking the status of running job $JOB_ID ..."
 for (( i=0; i<=9; i++ ))
 do  
     sleep 60
-    STAT=`kubectl get job  wiki-test --template={{.status.succeeded}}`
+    STAT=`kubectl get job  $JOB_ID --template={{.status.succeeded}}`
     if  [ "$STAT" == "<no value>" ]
     then
         echo "Seems to be in progress ..."

diff --git a/e2e/wikipedia-test.sh b/e2e/wikipedia-test.sh
@@ -21,6 +21,7 @@ done
 if [ $STATUS == "SUCCESS" ]
 then 
     echo "TASK $task_id COMPLETED SUCCESSFULLY"
+    sleep 60 # need time for the segments to become queryable
 else 
     echo "TASK $task_id  FAILED !!!!"
 fi