Merge pull request #49 from echen-98/maintenance-window

e2e tests to cover most replication group update scenarios

Author: kumargauravsharma

Diff for: test/e2e/bootstrap_resources.py

@@ -21,13 +21,17 @@
 
 @dataclass
 class TestBootstrapResources:
-    SnsTopicARN: str
-    SecurityGroupID: str
-    UserGroupID: str
+    SnsTopic1: str
+    SnsTopic2: str
+    SecurityGroup1: str
+    SecurityGroup2: str
+    UserGroup1: str
+    UserGroup2: str
     KmsKeyID: str
     SnapshotName: str
     NonDefaultUser: str
-    CWLogGroup: str
+    CWLogGroup1: str
+    CWLogGroup2: str
     CPGName: str
 
 _bootstrap_resources = None

Diff for: test/e2e/resources/replicationgroup_cme_misc.yaml

@@ -0,0 +1,16 @@
+# CME replication group with some optional fields specified. The modification of these fields is typically quicker
+# than other actions (such as scaling) which require provisioning/deletion of nodes.
+apiVersion: elasticache.services.k8s.aws/v1alpha1
+kind: ReplicationGroup
+metadata:
+  name: $RG_ID
+spec:
+  cacheNodeType: cache.t3.micro
+  engine: redis
+  numNodeGroups: 2
+  preferredMaintenanceWindow: $PMW
+  replicasPerNodeGroup: 1
+  replicationGroupDescription: $DESCRIPTION
+  replicationGroupID: $RG_ID
+  snapshotRetentionLimit: $SRL
+  snapshotWindow: $SW

Diff for: test/e2e/resources/replicationgroup_cme_ngc.yaml

@@ -0,0 +1,24 @@
+# CME replication group. "NGC" means that the field nodeGroupConfiguration is specified, with a detailed configuration
+# for each shard
+apiVersion: elasticache.services.k8s.aws/v1alpha1
+kind: ReplicationGroup
+metadata:
+  name: $RG_ID
+spec:
+  cacheNodeType: cache.t3.micro
+  engine: redis
+  nodeGroupConfiguration:
+    - nodeGroupID: $NGID1
+      primaryAvailabilityZone: us-west-2a
+      replicaAvailabilityZones:
+        - us-west-2b
+        - us-west-2c
+      replicaCount: 2
+    - nodeGroupID: $NGID2
+      primaryAvailabilityZone: us-west-2b
+      replicaAvailabilityZones:
+        - us-west-2c
+        - us-west-2a
+      replicaCount: 2
+  replicationGroupDescription: cluster-mode enabled RG
+  replicationGroupID: $RG_ID

Diff for: test/e2e/resources/replicationgroup_fault_tolerance.yaml

@@ -0,0 +1,14 @@
+# CMD replication group for testing modification fault tolerance features (multi AZ, auto failover, primary promotion)
+apiVersion: elasticache.services.k8s.aws/v1alpha1
+kind: ReplicationGroup
+metadata:
+  name: $RG_ID
+spec:
+  automaticFailoverEnabled: $AF_ENABLED
+  cacheNodeType: cache.t3.micro
+  engine: redis
+  multiAZEnabled: $MAZ_ENABLED
+  numNodeGroups: 1
+  replicasPerNodeGroup: 1
+  replicationGroupDescription: description
+  replicationGroupID: $RG_ID

Diff for: test/e2e/resources/replicationgroup_primary_cluster.yaml

@@ -0,0 +1,12 @@
+# CMD replication group manifest for specifying an existing node to use as primary
+apiVersion: elasticache.services.k8s.aws/v1alpha1
+kind: ReplicationGroup
+metadata:
+  name: $RG_ID
+spec:
+  # cache node type and engine cannot be used when specifying existing primary node, even if DNE
+  numNodeGroups: 1
+  replicasPerNodeGroup: 1
+  replicationGroupDescription: cluster-mode disabled RG
+  replicationGroupID: $RG_ID
+  primaryClusterID: $PRIMARY_NODE

Diff for: test/e2e/resources/replicationgroup_rpng.yaml

@@ -0,0 +1,15 @@
+# "RPNG" means that the field replicasPerNodeGroup is set (instead of nodeGroupConfiguration),
+# meaning that this replication group has a uniform configuration across all shards. This is one of the more
+# basic configurations as very few fields are specified and the resulting RG can either be CME or CMD
+# (depending on the number of specified node groups).
+apiVersion: elasticache.services.k8s.aws/v1alpha1
+kind: ReplicationGroup
+metadata:
+  name: $RG_ID
+spec:
+  cacheNodeType: cache.t3.micro
+  engine: redis
+  numNodeGroups: $NUM_NODE_GROUPS
+  replicasPerNodeGroup: $REPLICAS_PER_NODE_GROUP
+  replicationGroupDescription: cluster-mode enabled RG
+  replicationGroupID: $RG_ID

Diff for: test/e2e/service_bootstrap.py

@@ -136,12 +136,16 @@ def service_bootstrap() -> dict:
 
     return TestBootstrapResources(
         create_sns_topic(),
+        create_sns_topic(),
+        create_security_group(),
         create_security_group(),
         create_user_group(),
+        create_user_group(),
         create_kms_key(),
         create_cc_snapshot(),
         create_non_default_user(),
         create_log_group(),
+        create_log_group(),
         create_cpg()
     ).__dict__
 

Diff for: test/e2e/service_cleanup.py

@@ -86,19 +86,34 @@ def service_cleanup(config: dict):
     )
 
     try:
-        delete_sns_topic(resources.SnsTopicARN)
+        delete_sns_topic(resources.SnsTopic1)
     except:
-        logging.exception(f"Unable to delete SNS topic {resources.SnsTopicARN}")
+        logging.exception(f"Unable to delete SNS topic {resources.SnsTopic1}")
 
     try:
-        delete_security_group(resources.SecurityGroupID)
+        delete_sns_topic(resources.SnsTopic2)
     except:
-        logging.exception(f"Unable to delete VPC Security Group {resources.SecurityGroupID}")
+        logging.exception(f"Unable to delete SNS topic {resources.SnsTopic2}")
 
     try:
-        delete_user_group(resources.UserGroupID)
+        delete_security_group(resources.SecurityGroup1)
     except:
-        logging.exception(f"Unable to delete ElastiCache User Group {resources.UserGroupID}")
+        logging.exception(f"Unable to delete VPC Security Group {resources.SecurityGroup1}")
+
+    try:
+        delete_security_group(resources.SecurityGroup2)
+    except:
+        logging.exception(f"Unable to delete VPC Security Group {resources.SecurityGroup2}")
+
+    try:
+        delete_user_group(resources.UserGroup1)
+    except:
+        logging.exception(f"Unable to delete ElastiCache User Group {resources.UserGroup1}")
+
+    try:
+        delete_user_group(resources.UserGroup2)
+    except:
+        logging.exception(f"Unable to delete ElastiCache User Group {resources.UserGroup2}")
 
     try:
         delete_kms_key(resources.KmsKeyID)
@@ -116,9 +131,14 @@ def service_cleanup(config: dict):
         logging.exception(f"Unable to delete user {resources.NonDefaultUser}")
 
     try:
-        delete_log_group(resources.CWLogGroup)
+        delete_log_group(resources.CWLogGroup1)
+    except:
+        logging.exception(f"Unable to delete CW log group {resources.CWLogGroup1}")
+
+    try:
+        delete_log_group(resources.CWLogGroup2)
     except:
-        logging.exception(f"Unable to delete CW log group {resources.CWLogGroup}")
+        logging.exception(f"Unable to delete CW log group {resources.CWLogGroup2}")
 
     try:
         delete_cpg(resources.CPGName)