Merge branch 'lsst-sqre:main' into main

.pre-commit-config.yaml

@@ -14,7 +14,7 @@ repos:
           - -c=.yamllint.yml
 
   - repo: https://github.com/python-jsonschema/check-jsonschema
-    rev: 0.29.1
+    rev: 0.29.2
     hooks:
       - id: check-jsonschema
         files: ^applications/.*/secrets(-[^./-]+)?\.yaml
@@ -46,7 +46,7 @@ repos:
           - --template-files=../helm-docs.md.gotmpl
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.5.7
+    rev: v0.6.7
     hooks:
       - id: ruff
         args: [--fix, --exit-non-zero-on-fix]

Makefile

@@ -41,20 +41,20 @@ update-deps:
 	pip install --upgrade pip uv
 	uv pip install --upgrade pre-commit
 	pre-commit autoupdate
-	uv pip compile --upgrade --generate-hashes			\
-	    --output-file requirements/main.txt requirements/main.in
-	uv pip compile --upgrade --generate-hashes			\
+	uv pip compile --upgrade --universal --generate-hashes		\
+	    --output-file requirements/main.txt pyproject.toml
+	uv pip compile --upgrade --universal --generate-hashes		\
 	    --output-file requirements/dev.txt requirements/dev.in
-	uv pip compile --upgrade --generate-hashes			\
+	uv pip compile --upgrade --universal --generate-hashes		\
 	    --output-file requirements/tox.txt requirements/tox.in
 
 # Useful for testing against a Git version of Safir.
 .PHONY: update-deps-no-hashes
 update-deps-no-hashes:
 	pip install --upgrade uv
-	uv pip compile --upgrade					\
-	    --output-file requirements/main.txt requirements/main.in
-	uv pip compile --upgrade					\
+	uv pip compile --upgrade --universal				\
+	    --output-file requirements/main.txt pyproject.toml
+	uv pip compile --upgrade --universal				\
 	    --output-file requirements/dev.txt requirements/dev.in
-	uv pip compile --upgrade					\
+	uv pip compile --upgrade --universal				\
 	    --output-file requirements/tox.txt requirements/tox.in

applications/alert-stream-broker/README.md

@@ -72,12 +72,17 @@ Alert transmission to community brokers
 | alert-stream-broker.kafka.storage.size | string | `"1000Gi"` | Size of the backing storage disk for each of the Kafka brokers. |
 | alert-stream-broker.kafka.storage.storageClassName | string | `"standard"` | Name of a StorageClass to use when requesting persistent volumes. |
 | alert-stream-broker.kafka.version | string | `"3.4.0"` | Version of Kafka to deploy. |
+| alert-stream-broker.kafkaController.enabled | bool | `false` | Enable Kafka Controller |
+| alert-stream-broker.kafkaController.resources | object | See `values.yaml` | Kubernetes requests and limits for the Kafka Controller |
+| alert-stream-broker.kafkaController.storage.size | string | `"20Gi"` | Size of the backing storage disk for each of the Kafka controllers |
+| alert-stream-broker.kafkaController.storage.storageClassName | string | `""` | Name of a StorageClass to use when requesting persistent volumes |
 | alert-stream-broker.kafkaExporter | object | `{"enableSaramaLogging":false,"enabled":false,"groupRegex":".*","logLevel":"warning","topicRegex":".*"}` | Kafka JMX Exporter for more detailed diagnostic metrics. |
 | alert-stream-broker.kafkaExporter.enableSaramaLogging | bool | `false` | Enable Sarama logging |
 | alert-stream-broker.kafkaExporter.enabled | bool | `false` | Enable Kafka exporter. |
 | alert-stream-broker.kafkaExporter.groupRegex | string | `".*"` | Consumer groups to monitor |
 | alert-stream-broker.kafkaExporter.logLevel | string | `"warning"` | Log level for Sarama logging |
 | alert-stream-broker.kafkaExporter.topicRegex | string | `".*"` | Kafka topics to monitor |
+| alert-stream-broker.kraft | bool | `true` |  |
 | alert-stream-broker.maxBytesRetained | string | `"100000000000"` | Maximum number of bytes for the replay topic, per partition, per replica. Default is 100GB, but should be lower to not fill storage. |
 | alert-stream-broker.maxMillisecondsRetained | string | `"5259492000"` | Maximum amount of time to save alerts in the replay topic, in milliseconds. Default is 7 days (604800000). |
 | alert-stream-broker.nameOverride | string | `""` |  |
@@ -95,10 +100,8 @@ Alert transmission to community brokers
 | alert-stream-broker.users[0].readonlyTopics | list | `["alert-stream","alerts-simulated","alert-stream-test"]` | A list of topics that the user should get read-only access to. |
 | alert-stream-broker.users[0].username | string | `"rubin-testing"` | The username for the user that should be created. |
 | alert-stream-broker.vaultSecretsPath | string | `""` | Path to the secret resource in Vault |
-| alert-stream-broker.zookeeper.replicas | int | `3` | Number of Zookeeper replicas to run. |
-| alert-stream-broker.zookeeper.storage.size | string | `"1000Gi"` | Size of the backing storage disk for each of the Zookeeper instances. |
-| alert-stream-broker.zookeeper.storage.storageClassName | string | `"standard"` | Name of a StorageClass to use when requesting persistent volumes. |
 | alert-stream-schema-registry.clusterName | string | `"alert-broker"` | Strimzi "cluster name" of the broker to use as a backend. |
+| alert-stream-schema-registry.compatibilityLevel | string | `"None"` |  |
 | alert-stream-schema-registry.hostname | string | `"usdf-alert-schemas-dev.slac.stanford.edu"` | Hostname for an ingress which sends traffic to the Schema Registry. |
 | alert-stream-schema-registry.name | string | `"alert-schema-registry"` | Name used by the registry, and by its users. |
 | alert-stream-schema-registry.port | int | `8081` | Port where the registry is listening. NOTE: Not actually configurable in strimzi-registry-operator, so this basically cannot be changed. |

applications/alert-stream-broker/charts/alert-stream-broker/README.md

@@ -29,12 +29,17 @@ Kafka broker cluster for distributing alerts
 | kafka.storage.size | string | `"1000Gi"` | Size of the backing storage disk for each of the Kafka brokers. |
 | kafka.storage.storageClassName | string | `"standard"` | Name of a StorageClass to use when requesting persistent volumes. |
 | kafka.version | string | `"3.4.0"` | Version of Kafka to deploy. |
+| kafkaController.enabled | bool | `false` | Enable Kafka Controller |
+| kafkaController.resources | object | See `values.yaml` | Kubernetes requests and limits for the Kafka Controller |
+| kafkaController.storage.size | string | `"20Gi"` | Size of the backing storage disk for each of the Kafka controllers |
+| kafkaController.storage.storageClassName | string | `""` | Name of a StorageClass to use when requesting persistent volumes |
 | kafkaExporter | object | `{"enableSaramaLogging":false,"enabled":false,"groupRegex":".*","logLevel":"warning","topicRegex":".*"}` | Kafka JMX Exporter for more detailed diagnostic metrics. |
 | kafkaExporter.enableSaramaLogging | bool | `false` | Enable Sarama logging |
 | kafkaExporter.enabled | bool | `false` | Enable Kafka exporter. |
 | kafkaExporter.groupRegex | string | `".*"` | Consumer groups to monitor |
 | kafkaExporter.logLevel | string | `"warning"` | Log level for Sarama logging |
 | kafkaExporter.topicRegex | string | `".*"` | Kafka topics to monitor |
+| kraft | bool | `true` |  |
 | maxBytesRetained | string | `"100000000000"` | Maximum number of bytes for the replay topic, per partition, per replica. Default is 100GB, but should be lower to not fill storage. |
 | maxMillisecondsRetained | string | `"5259492000"` | Maximum amount of time to save alerts in the replay topic, in milliseconds. Default is 7 days (604800000). |
 | nameOverride | string | `""` |  |
@@ -52,6 +57,3 @@ Kafka broker cluster for distributing alerts
 | users[0].readonlyTopics | list | `["alert-stream","alerts-simulated","alert-stream-test"]` | A list of topics that the user should get read-only access to. |
 | users[0].username | string | `"rubin-testing"` | The username for the user that should be created. |
 | vaultSecretsPath | string | `""` | Path to the secret resource in Vault |
-| zookeeper.replicas | int | `3` | Number of Zookeeper replicas to run. |
-| zookeeper.storage.size | string | `"1000Gi"` | Size of the backing storage disk for each of the Zookeeper instances. |
-| zookeeper.storage.storageClassName | string | `"standard"` | Name of a StorageClass to use when requesting persistent volumes. |

applications/alert-stream-broker/charts/alert-stream-broker/templates/kafka-topics.yaml

@@ -26,6 +26,7 @@ spec:
     cleanup.policy: "delete"
     retention.ms: {{ .Values.maxMillisecondsRetained }} # 7 days
     retention.bytes: {{ .Values.maxBytesRetained }}
+    compression.type: {{ .Values.topicCompression}}
     # The default timestamp is the creation time of the alert.
     # To get the ingestion rate, we need this to be the log
     # append time, and the header will contain the producer
@@ -45,6 +46,7 @@ spec:
     cleanup.policy: "delete"
     retention.ms: {{ .Values.maxMillisecondsRetained }} # 7 days
     retention.bytes: {{ .Values.maxBytesRetained }}
+    compression.type: {{ .Values.topicCompression}}
     # The default timestamp is the creation time of the alert.
     # To get the ingestion rate, we need this to be the log
     # append time, and the header will contain the producer
@@ -64,11 +66,31 @@ spec:
     cleanup.policy: "delete"
     retention.ms: {{ .Values.maxMillisecondsRetained }} # 7 days
     retention.bytes: {{ .Values.maxBytesRetained }}
-    compression.type: {{ .Values.devTopicCompression}}
+    compression.type: {{ .Values.topicCompression}}
     # The default timestamp is the creation time of the alert.
     # To get the ingestion rate, we need this to be the log
     # append time, and the header will contain the producer
     # timestamp instead
     message.timestamp.type: 'LogAppendTime'
   partitions: {{ .Values.devTopicPartitions }}
-  replicas: {{ .Values.devTopicReplicas }}
+  replicas: {{ .Values.devTopicReplicas }}
+---
+apiVersion: "kafka.strimzi.io/{{ .Values.strimziAPIVersion }}"
+kind: KafkaTopic
+metadata:
+  labels:
+    strimzi.io/cluster: "{{ .Values.clusterName }}"
+  name: "{{ .Values.latissTopicName}}"
+spec:
+  config:
+    cleanup.policy: "delete"
+    retention.ms: {{ .Values.maxMillisecondsRetained }} # 7 days
+    retention.bytes: {{ .Values.maxBytesRetained }}
+    compression.type: {{ .Values.topicCompression}}
+    # The default timestamp is the creation time of the alert.
+    # To get the ingestion rate, we need this to be the log
+    # append time, and the header will contain the producer
+    # timestamp instead
+    message.timestamp.type: 'LogAppendTime'
+  partitions: {{ .Values.latissTopicPartitions }}
+  replicas: {{ .Values.latissTopicReplicas }}

applications/alert-stream-broker/charts/alert-stream-broker/templates/kafka.yaml

@@ -1,7 +1,60 @@
+apiVersion: kafka.strimzi.io/v1beta2
+kind: KafkaNodePool
+metadata:
+  name: controller
+  labels:
+    strimzi.io/cluster: {{ .Values.cluster.name }}
+spec:
+  replicas: {{ .Values.kafka.replicas }}
+  roles:
+    - controller
+  storage:
+    type: jbod
+    volumes:
+    - id: 0
+      type: persistent-claim
+      size: {{ .Values.kafkaController.storage.size }}
+      class: {{ .Values.kafkaController.storage.storageClassName }}
+      deleteClaim: false
+  {{- with .Values.kafkaController.resources }}
+  resources:
+    {{- toYaml . | nindent 6 }}
+  {{- end }}
+---
+apiVersion: kafka.strimzi.io/v1beta2
+kind: KafkaNodePool
+metadata:
+  name: kafka
+  labels:
+    strimzi.io/cluster: {{ .Values.cluster.name }}
+  annotations:
+    strimzi.io/next-node-ids: "[0-99]"
+spec:
+  replicas: {{ .Values.kafka.replicas }}
+  roles:
+    - broker
+  storage:
+    type: jbod
+    volumes:
+    - id: 0
+      type: persistent-claim
+      size: {{ .Values.kafka.storage.size }}
+      {{- if .Values.kafka.storage.storageClassName }}
+      class: {{ .Values.kafka.storage.storageClassName }}
+      {{- end}}
+      deleteClaim: false
+  {{- with .Values.kafka.resources }}
+  resources:
+    {{- toYaml . | nindent 6 }}
+  {{- end }}
+---
 apiVersion: kafka.strimzi.io/{{ .Values.strimziAPIVersion }}
 kind: Kafka
 metadata:
   name: {{ .Values.cluster.name }}
+  annotations:
+    strimzi.io/kraft: enabled
+    strimzi.io/node-pools: enabled
 spec:
   {{- if .Values.kafkaExporter.enabled }}
   kafkaExporter:
@@ -85,14 +138,15 @@ spec:
 
           {{- if .Values.kafka.externalListener.brokers }}
           brokers:
-            {{- range $idx, $broker := .Values.kafka.externalListener.brokers }}
-            - broker: {{ $idx }}
+          {{- range $broker := .Values.kafka.externalListener.brokers }}
+            - broker: {{ $broker.broker }}
               loadBalancerIP: {{ $broker.ip }}
               advertisedHost: {{ $broker.host }}
-              annotations: {{ toYaml $broker.annotations | nindent 16 }}
-            {{- end }}
+              advertisedPort: 9094
+              annotations:
+                annotations: {{ toYaml $broker.annotations | nindent 16 }}
+          {{- end }}
           {{- end }}
-
           {{- if and (.Values.kafka.externalListener.tls.enabled) (.Values.kafka.externalListener.bootstrap.host) }}
           brokerCertChainAndKey:
             secretName: {{ .Values.cluster.name }}-external-tls
@@ -114,8 +168,6 @@ spec:
       transaction.state.log.replication.factor: 3
       transaction.state.log.min.isr: 2
       message.max.bytes: 4194304 # 8 Megabytes. For testing purposes only.
-      log.message.format.version: {{ .Values.kafka.logMessageFormatVersion }}
-      inter.broker.protocol.version: {{ .Values.kafka.interBrokerProtocolVersion }}
       ssl.client.auth: required
       {{- range $key, $value := .Values.kafka.config }}
       {{ $key }}: {{ $value }}
@@ -133,53 +185,6 @@ spec:
         class: {{ .Values.kafka.storage.storageClassName }}
         deleteClaim: false
 
-    template:
-      pod:
-        {{- if .Values.kafka.nodePool.tolerations }}
-        tolerations:
-          {{- range $tol := .Values.kafka.nodePool.tolerations }}
-          - key: {{ $tol.key }}
-            operator: "Equal"
-            value: {{ $tol.value }}
-            effect: {{ $tol.effect }}
-          {{- end }}
-        {{- end }}
-
-        {{- if .Values.kafka.nodePool.affinities }}
-        affinity:
-          nodeAffinity:
-            preferredDuringSchedulingIgnoredDuringExecution:
-              {{- range $affinity := .Values.kafka.nodePool.affinities }}
-              - weight: 1
-                preference:
-                  matchExpressions:
-                      - key: {{ $affinity.key }}
-                        operator: In
-                        values: [{{ $affinity.value }}]
-              {{- end }}
-        {{- end }}
-
-        affinity:
-          podAntiAffinity:
-            requiredDuringSchedulingIgnoredDuringExecution:
-            - labelSelector:
-                matchExpressions:
-                  - key:  "app.kubernetes.io/name"
-                    operator: In
-                    values:
-                    - kafka
-              topologyKey: "kubernetes.io/hostname"
-
-  zookeeper:
-    replicas: {{ .Values.zookeeper.replicas }}
-    storage:
-      # Note that storage is configured per replica. If there are 3 replicas,
-      # each will get its own PersistentVolumeClaim for the configured size.
-      type: persistent-claim
-      size: {{ .Values.zookeeper.storage.size }}
-      class: {{ .Values.zookeeper.storage.storageClassName }}
-      deleteClaim: false
-
     template:
       pod:
         {{- if .Values.kafka.nodePool.tolerations }}

applications/alert-stream-broker/charts/alert-stream-broker/values.yaml

@@ -121,17 +121,6 @@ users:
     # matches.
     groups: ["rubin-testing"]
 
-
-zookeeper:
-  # -- Number of Zookeeper replicas to run.
-  replicas: 3
-
-  storage:
-    # -- Size of the backing storage disk for each of the Zookeeper instances.
-    size: 1000Gi
-    # -- Name of a StorageClass to use when requesting persistent volumes.
-    storageClassName: standard
-
 tls:
   subject:
     # -- Organization to use in the 'Subject' field of the broker's TLS certificate.
@@ -149,6 +138,29 @@ fullnameOverride: ""
 
 nameOverride: ""
 
+kraft: true
+
+kafkaController:
+  # -- Enable Kafka Controller
+  enabled: false
+
+  storage:
+    # -- Size of the backing storage disk for each of the Kafka controllers
+    size: 20Gi
+
+    # -- Name of a StorageClass to use when requesting persistent volumes
+    storageClassName: ""
+
+  # -- Kubernetes requests and limits for the Kafka Controller
+  # @default -- See `values.yaml`
+  resources:
+    requests:
+      memory: 32Gi
+      cpu: "4"
+    limits:
+      memory: 64Gi
+      cpu: "8"
+
 # -- Topic used to send test alerts.
 testTopicName: alert-stream-test
 

applications/alert-stream-broker/charts/alert-stream-schema-registry/README.md

@@ -7,6 +7,7 @@ Confluent Schema Registry for managing schema versions for the Alert Stream
 | Key | Type | Default | Description |
 |-----|------|---------|-------------|
 | clusterName | string | `"alert-broker"` | Strimzi "cluster name" of the broker to use as a backend. |
+| compatibilityLevel | string | `"None"` |  |
 | hostname | string | `"usdf-alert-schemas-dev.slac.stanford.edu"` | Hostname for an ingress which sends traffic to the Schema Registry. |
 | name | string | `"alert-schema-registry"` | Name used by the registry, and by its users. |
 | port | int | `8081` | Port where the registry is listening. NOTE: Not actually configurable in strimzi-registry-operator, so this basically cannot be changed. |

applications/alert-stream-broker/charts/alert-stream-schema-registry/templates/schema-registry-server.yaml

@@ -9,4 +9,5 @@ metadata:
     revision: "1"
 spec:
   strimzi-version: {{ .Values.strimziAPIVersion }}
-  listener: internal
+  listener: internal
+  compatibilityLevel: none