chore: Add 35.0.1, remove 33.0.0 (#786)

Techassi · web-flow · commit e324ed4324b2 · 2026-01-23T10:55:19.000Z
* docs: Update supported versions for 26.3

* test: Add 35.0.1, remove 33.0.0 from test definition

* chore: Update Druid version from 34.0.0 to 35.0.1

* chore: Bump CPU request and limit of router container

Since Druid 35.0.1 the router container failed to start up due to the
following error: 'Insufficient configured threads'. The CPU resources
are now bumped to be in line with other containers, like the historical
and broker containers.

* test: Correct URL due to TLS SNI errors

* chore: Bump examples/test resources to 35.0.1

* docs: Update resource requirements

* docs: Update comment about resource calculation
diff --git a/docs/modules/druid/examples/getting_started/druid.yaml b/docs/modules/druid/examples/getting_started/druid.yaml
@@ -5,7 +5,7 @@ metadata:
   name: simple-druid
 spec:
   image:
-    productVersion: 34.0.0
+    productVersion: 35.0.1
   clusterConfig:
     zookeeperConfigMapName: simple-druid-znode
     deepStorage:
diff --git a/docs/modules/druid/examples/getting_started/druid.yaml.j2 b/docs/modules/druid/examples/getting_started/druid.yaml.j2
@@ -5,7 +5,7 @@ metadata:
   name: simple-druid
 spec:
   image:
-    productVersion: 34.0.0
+    productVersion: 35.0.1
   clusterConfig:
     zookeeperConfigMapName: simple-druid-znode
     deepStorage:
diff --git a/docs/modules/druid/pages/usage-guide/resources-and-storage.adoc b/docs/modules/druid/pages/usage-guide/resources-and-storage.adoc
@@ -11,8 +11,11 @@ include::home:concepts:stackable_resource_requests.adoc[]
 
 A minimal HA setup consisting of 2 Pods of each role has the following https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/[resource requirements]:
 
-* `4700m` CPU request
-* `13800m` CPU limit
+// We are not sure how these values were calculated in the first place, but we assume the are calculated like this:
+// The sum of resources listed below + HDFS resources + ZK resources + some overhead (maybe for operators).
+// The amount of resources for HDFS and ZK are taken from their individual documentation pages.
+* `5100m` CPU request
+* `15600m` CPU limit
 * `12144Mi` memory request and limit
 
 Of course, additional services require additional resources.
@@ -35,8 +38,8 @@ spec:
     config:
       resources:
         cpu:
-          min: 100m
-          max: 400m
+          min: 300m
+          max: 1200m
         memory:
           limit: 512Mi
   historical:
diff --git a/docs/modules/druid/partials/supported-versions.adoc b/docs/modules/druid/partials/supported-versions.adoc
@@ -2,6 +2,6 @@
 // This is a separate file, since it is used by both the direct Druid documentation, and the overarching
 // Stackable Platform documentation.
 
-- 34.0.0
-- 33.0.0 (deprecated)
+- 35.0.1
+- 34.0.0 (deprecated)
 - 30.0.1 (LTS)
diff --git a/examples/psql-s3/psql-s3-druid-cluster.yaml b/examples/psql-s3/psql-s3-druid-cluster.yaml
@@ -45,7 +45,7 @@ metadata:
   name: psql-s3-druid
 spec:
   image:
-    productVersion: 31.0.1
+    productVersion: 35.0.1
   clusterConfig:
     deepStorage:
       s3:
diff --git a/examples/psql/psql-hdfs-druid-cluster.yaml b/examples/psql/psql-hdfs-druid-cluster.yaml
@@ -56,7 +56,7 @@ metadata:
   name: psql-druid
 spec:
   image:
-    productVersion: 31.0.1
+    productVersion: 35.0.1
   clusterConfig:
     deepStorage:
       hdfs:
diff --git a/examples/tls/tls-druid-cluster.yaml b/examples/tls/tls-druid-cluster.yaml
@@ -78,7 +78,7 @@ metadata:
   name: derby-druid
 spec:
   image:
-    productVersion: 31.0.1
+    productVersion: 35.0.1
   clusterConfig:
     authentication:
       - authenticationClass: druid-mtls-authentication-class
diff --git a/rust/operator-binary/src/crd/resource.rs b/rust/operator-binary/src/crd/resource.rs
@@ -232,8 +232,8 @@ pub static COORDINATOR_RESOURCES: LazyLock<
 pub static ROUTER_RESOURCES: LazyLock<ResourcesFragment<storage::DruidStorage, NoRuntimeLimits>> =
     LazyLock::new(|| ResourcesFragment {
         cpu: CpuLimitsFragment {
-            min: Some(Quantity("100m".to_owned())),
-            max: Some(Quantity("400m".to_owned())),
+            min: Some(Quantity("300m".to_owned())),
+            max: Some(Quantity("1200m".to_owned())),
         },
         memory: MemoryLimitsFragment {
             limit: Some(Quantity("512Mi".to_owned())),
diff --git a/rust/operator-binary/test/resources/crd/resource_merge/druid_cluster.yaml b/rust/operator-binary/test/resources/crd/resource_merge/druid_cluster.yaml
@@ -7,7 +7,7 @@ metadata:
   uid: test-resource-merge
 spec:
   image:
-    productVersion: 30.0.0
+    productVersion: 35.0.1
   clusterConfig:
     deepStorage:
       hdfs:
diff --git a/rust/operator-binary/test/resources/crd/resource_merge/segment_cache.yaml b/rust/operator-binary/test/resources/crd/resource_merge/segment_cache.yaml
@@ -7,7 +7,7 @@ metadata:
   uid: test-resource-merge
 spec:
   image:
-    productVersion: 30.0.0
+    productVersion: 35.0.1
   clusterConfig:
     deepStorage:
       hdfs:
diff --git a/rust/operator-binary/test/resources/crd/role_service/druid_cluster.yaml b/rust/operator-binary/test/resources/crd/role_service/druid_cluster.yaml
@@ -7,7 +7,7 @@ metadata:
   uid: test-uid
 spec:
   image:
-    productVersion: 30.0.0
+    productVersion: 35.0.1
   clusterConfig:
     deepStorage:
       hdfs:
diff --git a/rust/operator-binary/test/resources/druid_controller/segment_cache.yaml b/rust/operator-binary/test/resources/druid_controller/segment_cache.yaml
@@ -7,7 +7,7 @@ metadata:
   uid: test-resource-merge
 spec:
   image:
-    productVersion: 30.0.0
+    productVersion: 35.0.1
   clusterConfig:
     deepStorage:
       s3:
diff --git a/rust/operator-binary/test/resources/druid_controller/simple.yaml b/rust/operator-binary/test/resources/druid_controller/simple.yaml
@@ -7,7 +7,7 @@ metadata:
   uid: test-uid
 spec:
   image:
-    productVersion: 30.0.0
+    productVersion: 35.0.1
   clusterConfig:
     deepStorage:
       hdfs:
diff --git a/tests/templates/kuttl/commons/healthcheck.py b/tests/templates/kuttl/commons/healthcheck.py
@@ -14,6 +14,7 @@
     )
 
     druid_cluster_name = sys.argv[1]
+    namespace = sys.argv[2]
 
     druid_role_ports = {
         "broker": 8282,
@@ -24,9 +25,7 @@
     }
 
     for role, port in druid_role_ports.items():
-        url = (
-            f"https://{druid_cluster_name}-{role}-default-headless:{port}/status/health"
-        )
+        url = f"https://{druid_cluster_name}-{role}-default-headless.{namespace}.svc.cluster.local:{port}/status/health"
         count = 1
 
         # As this script is intended to be executed by Kuttl which is in charge of overall test timeouts it is ok
diff --git a/tests/templates/kuttl/hdfs-deep-storage/05-assert.yaml b/tests/templates/kuttl/hdfs-deep-storage/05-assert.yaml
@@ -2,5 +2,5 @@
 apiVersion: kuttl.dev/v1beta1
 kind: TestAssert
 commands:
-  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py derby-druid
+  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py derby-druid $NAMESPACE
 timeout: 300
diff --git a/tests/templates/kuttl/ingestion-no-s3-ext/05-assert.yaml b/tests/templates/kuttl/ingestion-no-s3-ext/05-assert.yaml
@@ -2,5 +2,5 @@
 apiVersion: kuttl.dev/v1beta1
 kind: TestAssert
 commands:
-  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py derby-druid
+  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py derby-druid $NAMESPACE
 timeout: 300
diff --git a/tests/templates/kuttl/ingestion-s3-ext/05-assert.yaml b/tests/templates/kuttl/ingestion-s3-ext/05-assert.yaml
@@ -2,5 +2,5 @@
 apiVersion: kuttl.dev/v1beta1
 kind: TestAssert
 commands:
-  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py derby-druid
+  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py derby-druid $NAMESPACE
 timeout: 300
diff --git a/tests/templates/kuttl/s3-deep-storage/11-assert.yaml b/tests/templates/kuttl/s3-deep-storage/11-assert.yaml
@@ -2,5 +2,5 @@
 apiVersion: kuttl.dev/v1beta1
 kind: TestAssert
 commands:
-  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py druid-s3-deep-storage
+  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py druid-s3-deep-storage $NAMESPACE
     timeout: 300
diff --git a/tests/templates/kuttl/smoke/70-assert.yaml b/tests/templates/kuttl/smoke/70-assert.yaml
@@ -2,5 +2,5 @@
 apiVersion: kuttl.dev/v1beta1
 kind: TestAssert
 commands:
-  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py druid
+  - script: kubectl exec -n $NAMESPACE checks-0 -- python /tmp/healthcheck.py druid $NAMESPACE
 timeout: 300
diff --git a/tests/test-definition.yaml b/tests/test-definition.yaml
@@ -14,13 +14,13 @@ dimensions:
   - name: druid
     values:
       - 30.0.1
-      - 33.0.0
       - 34.0.0
+      - 35.0.1
       # To use a custom image, add a comma and the full name after the product version
       # - 30.0.0,oci.stackable.tech/sdp/druid:30.0.0-stackable0.0.0-dev
   - name: druid-latest
     values:
-      - 34.0.0
+      - 35.0.1
       # To use a custom image, add a comma and the full name after the product version
       # - 30.0.0,oci.stackable.tech/sdp/druid:30.0.0-stackable0.0.0-dev
   - name: zookeeper
