Skip to content

Latest commit

 

History

History
538 lines (397 loc) · 24 KB

CHANGELOG.md

File metadata and controls

538 lines (397 loc) · 24 KB

Changelog

All notable changes to this project will be documented in this file.

[Unreleased]

Added

Changed

  • Bump stackable-operator to 0.108.0 and strum to 0.28 (#943, #945).
  • Gracefully shutdown all concurrent tasks by forwarding the SIGTERM signal (#934).
  • Refactor: move server configuration properties from the command line to configuration files. (#911).
  • Add support for ZooKeeper to KRaft migration (#923), (#933).
  • Bump testing-tools to 0.3.0-stackable0.0.0-dev (#925).

Fixed

  • Fix "404 page not found" error for the initial object list (#945).

Removed

  • Refactor: remove unused RBAC cluster role (#914).
  • Refactor: remove superfluous and partly misconfigured Kafka listeners CLIENT_AUTH and CONTROLLER_AUTH (#915).
  • Remove support for 3.7.2 and 4.1.0 (#929)

[25.11.0] - 2025-11-07

[25.11.0-rc1] - 2025-11-06

Added

  • Helm: Allow Pod priorityClassName to be configured (#890).

  • Add experimental support for Kafka KRaft mode (#889).

  • Add experimental support for Kafka 4.1.0 (#889).

  • Add prometheus.io/path|port|scheme annotations to metrics service (#897).

  • Add client.properties to the Kafka configuration config map (#898).

    Use this file together with the Kafka client shell scripts and preconfigured TLS settings. Unfortunately, when Kerberos is enabled this file is incomplete and must be edited first before it can be used.

  • Add end-of-support checker (#895).

    • EOS_CHECK_MODE (--eos-check-mode) to set the EoS check mode. Currently, only "offline" is supported.
    • EOS_INTERVAL (--eos-interval) to set the interval in which the operator checks if it is EoS.
    • EOS_DISABLED (--eos-disabled) to disable the EoS checker completely.

Changed

  • Deprecate support for Kafka 3.7.2 (#892).
  • BREAKING: The <cluster>-<role>-<rolegroup> rolegroup service was replaced with a <cluster>-<role>-<rolegroup>-headless and <cluster>-<role>-<rolegroup>-metrics rolegroup service (#897).
  • Small cleanups and updates (#900)
    • remove the metrics port from services that don't need it
    • use the new server.yaml for jmx configuration
    • update metrics tests
    • update monitoring doc
  • Replace kcat with Kafka client scripts wherever possible (#898). At the moment, kcat is still used for liveliness probes and Kerberos tests.
  • Bump stackable-operator to 0.100.1 (#906).

[25.7.0] - 2025-07-23

[25.7.0-rc1] - 2025-07-18

Added

  • Adds new telemetry CLI arguments and environment variables (#851).
    • Use --file-log-max-files (or FILE_LOG_MAX_FILES) to limit the number of log files kept.
    • Use --file-log-rotation-period (or FILE_LOG_ROTATION_PERIOD) to configure the frequency of rotation.
    • Use --console-log-format (or CONSOLE_LOG_FORMAT) to set the format to plain (default) or json.
  • Add support for Kafka 3.9.1 (#866).
  • Add RBAC rule to helm template for automatic cluster domain detection (#878).

Changed

  • BREAKING: Replace stackable-operator initialize_logging with stackable-telemetry Tracing (#840, #847, #851).
    • The console log level was set by KAFKA_OPERATOR_LOG, and is now set by CONSOLE_LOG_LEVEL.
    • The file log level was set by KAFKA_OPERATOR_LOG, and is now set by FILE_LOG_LEVEL.
    • The file log directory was set by KAFKA_OPERATOR_LOG_DIRECTORY, and is now set by FILE_LOG_DIRECTORY (or via --file-log-directory <DIRECTORY>).
    • Replace stackable-operator print_startup_string with tracing::info! with fields.
  • BREAKING: Inject the vector aggregator address into the vector config using the env var VECTOR_AGGREGATOR_ADDRESS instead of having the operator write it to the vector config (#844).
  • test: Bump to Vector 0.46.1 (#855).
  • Use versioned common structs (#861).
  • BREAKING: Previously this operator would hardcode the UID and GID of the Pods being created to 1000/0, this has changed now (#862)
    • The runAsUser and runAsGroup fields will not be set anymore by the operator
    • The defaults from the docker images itself will now apply, which will be different from 1000/0 going forward
    • This is marked as breaking because tools and policies might exist, which require these fields to be set
  • BREAKING: Bump stackable-operator to 0.94.0 and update other dependencies (#878).
    • The default Kubernetes cluster domain name is now fetched from the kubelet API unless explicitly configured.
    • This requires operators to have the RBAC permission to get nodes/proxy in the apiGroup "". The helm-chart takes care of this.
    • The CLI argument --kubernetes-node-name or env variable KUBERNETES_NODE_NAME needs to be set. The helm-chart takes care of this.
  • The operator helm-chart now grants RBAC patch permissions on events.k8s.io/events, so events can be aggregated (e.g. "error happened 10 times over the last 5 minutes") (#879).

Fixed

  • Use json file extension for log files (#846).
  • Fix a bug where changes to ConfigMaps that are referenced in the KafkaCluster spec didn't trigger a reconciliation (#844).
  • Allow uppercase characters in domain names (#878).

Removed

  • test: ZooKeeper 3.9.2 removed (#853).
  • Support for Kafka 3.7.1 and 3.8.0 removed (#860).
  • Remove the -nodeport discovery ConfigMap (#868).
  • Remove the lastUpdateTime field from the stacklet status (#878).
  • Remove role binding to legacy service accounts (#878).

[25.3.0] - 2025-03-21

Deprecated

  • The -nodeport discovery ConfigMaps have been deprecated for removal (#813).
    • Any existing uses should be migrated to the primary discovery ConfigMap instead.

Added

  • The lifetime of auto generated TLS certificates is now configurable with the role and roleGroup config property requestedSecretLifetime. This helps reduce frequent Pod restarts (#796).
  • Run a containerdebug process in the background of each Kafka container to collect debugging information (#803).
  • Aggregate emitted Kubernetes events on the CustomResources (#809).
  • Support configuring JVM arguments (#819).
  • Add Kafka 3.9.0 to tests (#822).

Changed

  • Bump stackable-operator to 0.87.0 and stackable-versioned to 0.6.0 (#830).
  • Default to OCI for image metadata and product image selection (#810).
  • Bump Kafka 3.7.1 to 3.7.2 in tests and getting_started, and bump upgrade testing from 3.7.1->3.8.0 to 3.8.0->3.9.0 (#822).
  • docs: Update supported versions list (#835).

Fixed

  • Readiness probe fixed if Kerberos is enabled (#833).

[24.11.1] - 2025-01-10

Fixed

  • BREAKING: Use distinct ServiceAccounts for the Stacklets, so that multiple Stacklets can be deployed in one namespace. Existing Stacklets will use the newly created ServiceAccounts after restart (#793).

[24.11.0] - 2024-11-18

Added

  • Support version 3.8.0 (#753).
  • Add support for Kerberos authentication (#762).
  • The operator can now run on Kubernetes clusters using a non-default cluster domain. Use the env var KUBERNETES_CLUSTER_DOMAIN or the operator Helm chart property kubernetesClusterDomain to set a non-default cluster domain (#771).

Changed

  • Reduce CRD size from 479KB to 53KB by accepting arbitrary YAML input instead of the underlying schema for the following fields (#750):
    • podOverrides
    • affinity
  • Migrate to exposing Kafka using Listener Operator (#443).
    • BREAKING: The existing services will be migrated to the new format. Clients will need to re-read settings from the discovery configmap.
    • BREAKING: Kafka is now only accessible from within the Kubernetes cluster by default. Set listener classes manually to expose it to the outside world (again).
    • BREAKING: To complete an upgrade to this kafka-operator, all existing Kafka StatefulSets must be deleted manually. This will cause some downtime.

Fixed

  • Include the global Kafka bootstrap service (not the rolegroup-specific) DNS record as SAN entry in the generated certificates used by Kafka. This allows you to access Kafka brokers secured using TLS via the global bootstrap service (#741).
  • An invalid KafkaCluster doesn't cause the operator to stop functioning (#773).

Removed

  • Remove versions 3.4.1, 3.6.1, 3.6.2 (#753).

[24.7.0] - 2024-07-24

Added

  • Support for versions 3.6.2, 3.7.1 (#723).

Changed

  • Bump stackable-operator from 0.64.0 to 0.70.0 (#725).
  • Bump product-config from 0.6.0 to 0.7.0 (#725).
  • Bump other dependencies (#728).

Removed

  • Support for version 3.5.2 (#723).
  • BREAKING: Remove field/arg controller_config from kafka_controller::Ctx struct and create_controller function (#726).

[24.3.0] - 2024-03-20

Added

  • Various documentation of the CRD (#645).
  • Helm: support labels in values.yaml (#657).
  • Support new versions 3.5.2, 3.6.1 (#664).

Removed

  • Support for versions 2.8.2, 3.4.0, 3.5.1 (#664).

Fixed

  • Processing of corrupted log events fixed; If errors occur, the error messages are added to the log event (#715).

[23.11.0] - 2023-11-24

Added

  • Default stackableVersion to operator version. It is recommended to remove spec.image.stackableVersion from your custom resources (#611, #613).
  • Configuration overrides for the JVM security properties, such as DNS caching (#616).
  • Support PodDisruptionBudgets (#625).
  • Support new versions 2.8.2, 3.4.1, 3.5.1 (#627).
  • Document internal clusterId check (#631).
  • Support graceful shutdown (#635).

Changed

  • vector 0.26.0 -> 0.33.0 (#612, #627).
  • operator-rs 0.44.0 -> 0.55.0 (#611, #621, #625, #627).
  • [BREAKING]: Let secret-operator handle certificate conversion. Doing so we were able to remove the prepare init container with the effect, that you can't configure the log level for this container anymore. You need to remove the field spec.brokers.config.logging.container.prepare in case you have specified it (#621).
  • Combine the operator lib and binary crates (#638).

Removed

  • Removed support for versions 2.7.1, 3.1.0, 3.2.0, 3.3.1 (#627).

[23.7.0] - 2023-07-14

Added

  • Generate OLM bundle for Release 23.4.0 (#585).
  • Fixed upgrade test on Openshift (#585).
  • Missing CRD defaults for status.conditions field (#588).
  • Support Kafka 3.4.0 (#591).
  • Add support for resource quotas (#595)
  • Support podOverrides (#602)

Fixed

  • Increase the size limit of the log volume (#604)

Changed

  • operator-rs 0.40.2 -> 0.44.0 (#583, #604).
  • Use 0.0.0-dev product images for testing (#584)
  • Use testing-tools 0.2.0 (#584)
  • Added kuttl test suites (#599)

[23.4.0] - 2023-04-17

Added

  • Enabled logging and log aggregation (#547).
  • Deploy default and support custom affinities (#557).
  • Openshift compatibility (#572).
  • Extend cluster resources for status and cluster operation (paused, stopped) (#574).
  • Cluster status conditions (#575).

Changed

  • operator-rs 0.30.1 -> 0.40.2 (#545, #572, #574, #577).
  • Bumped stackable versions to "23.4.0-rc1" (#545).
  • Bumped kafka stackable versions to "23.4.0-rc2" (#547).
  • Use operator-rs build_rbac_resources method (#572).
  • Updated landing page and restructured usage guide (#573).

Fixed

  • Avoid empty log events dated to 1970-01-01 and improve the precision of the log event timestamps (#577).

[23.1.0] - 2023-01-23

Changed

  • Fixed the RoleGroup selector. It was not used before. (#530)
  • Updated stackable image versions (#513).
  • operator-rs: 0.26.0 -> 0.30.1 (#519, #530).
  • Don't run init container as root and avoid chmod and chowning (#524).
  • [BREAKING] Use Product image selection instead of version. spec.version has been replaced by spec.image (#482).
  • [BREAKING]: Removed tools image for init and get-svc container and replaced with Kafka product image. This means the latest stackable version has to be used in the product image selection (#527)
  • [BREAKING] Consolidated top-level configuration. Split up TLS encryption and authentication. Moved all top-level fields except spec.image below spec.clusterConfig (#532).

[0.8.0] - 2022-11-07

Added

  • Added default resource requests (memory and cpu) for Kafka pods (#485).
  • Support for Kafka 3.3.1 (#492).
  • Orphaned resources are deleted (#495).

Changed

  • Change port names from http/https to kafka/kafka-tls (#472).
  • Role and rolegroup configurations are merged correctly (#499).
  • operator-rs: 0.22.0 -> 0.26.0 (#495, #499).

[0.7.0] - 2022-09-06

Added

  • BREAKING: TLS encryption and authentication support for internal and client communications. This is breaking for clients because the cluster is secured per default, which results in a client port change (#442).

Changed

  • operator-rs: 0.21.1 -> 0.22.0 (#430).
  • Include chart name when installing with a custom release name (#429, #431).
  • Kafka init container now uses Stackable tools rather than Bitnami kubectl (#434).

[0.6.0] - 2022-06-30

Added

  • Reconciliation errors are now reported as Kubernetes events (#346).
  • Support for Kafka 3.1.0 (#347).
  • Use cli argument watch-namespace / env var WATCH_NAMESPACE to specify a single namespace to watch (#351).
  • Optional CRD field log4j to adapt the log4j.properties (#364).
  • PVCs for data storage, cpu and memory limits are now configurable (#405).
  • Moved tests from integration tests repo to operator repo (#409).

Changed

  • operator-rs 0.10.0 → 0.21.1 (#346, #351, #385, #405).
  • --kafka-broker-clusterrole is now only accepted for the run subcommand (#349).
  • BREAKING: Adapted the opa field in the crd to opaConfigMapName and fixed authorizer.class.name to org.openpolicyagent.kafka.OpaAuthorizer and opa.authorizer.metrics.enabled to true. Other settings can be changed via configOverrides (#364).
  • BREAKING: opaConfigMapName in CRD adapted to opa using the OpaConfig from operator-rs (#385).
  • BREAKING: Specifying the product version has been changed to adhere to ADR018 instead of just specifying the product version you will now have to add the Stackable image version as well, so version: 3.1.0 becomes (for example) version: 3.1.0-stackable0 (#409)

[0.5.0] - 2022-02-14

Changed

  • Complete rework (#256).

[0.4.0] - 2021-12-06

  • operator-rs 0.3.0 → 0.4.0 (#214).
  • stackable-opa-crd 0.4.1 → 0.5.0 (#214).
  • stackable-zookeeper-crd 0.4.1 → 0.5.0 (#214).
  • Adapted pod image and container command to docker image (#214).
  • Adapted documentation to represent new workflow with docker images (#214).

[0.3.0] - 2021-10-27

Added

  • Added versioning code from operator-rs for up and downgrades (#167).
  • Added ProductVersion to status (#167).
  • Added Condition to status (#167).
  • Use sticky scheduler (#181)
  • Added support for Start, Stop Restart commands (#194).

Changed

  • operator-rs 0.2.2 → 0.3.0 (#207).
  • stackable-zookeeper-crd: 0.2.0 → 0.4.1 (#207).
  • stackable-opa-crd: 0.2.0 → 0.4.1 (#207).
  • kube-rs: 0.58 → 0.60 (#167).
  • k8s-openapi 0.12 → 0.13 and features: v1_21 → v1_22 (#167).
  • stackable-zookeeper-crd::util to stackable-zookeeper-crd::discovery (#194).
  • Moved CRD availability check to operator-binary (#194).

Removed

  • kube-runtime dependency (#167).

[0.2.1] - 2021-09-14

  • Fixed Dockerfile to use the correct binary (#167).

[0.2.0] - 2021.09.10

Changed

  • Breaking: Repository structure was changed and the -server crate renamed to -binary. As part of this change the -server suffix was removed from both the package name for os packages and the name of the executable (#157).

[0.1.0] - 2021.09.07

Added

  • Initial release