Skip to content

CHANGELOG.md

Latest commit

 

History

History
500 lines (359 loc) · 22.1 KB

CHANGELOG.md

File metadata and controls

500 lines (359 loc) · 22.1 KB

Changelog

[Unreleased]

Changed

  • Document Helm deployed RBAC permissions and remove unnecessary permissions (#717).
  • BREAKING: configOverrides now only accepts the supported config file name superset_config.py. Previously arbitrary keys were silently accepted but ignored (#719).
  • Bump stackable-operator to 0.110.0 and kube to 3.1.0 (#719).
  • Support setting clientAuthenticationMethod for OIDC authentication. The value is passed through to the Flask-AppBuilder config as token_endpoint_auth_method (#719).

[26.3.0] - 2026-03-16

[26.3.0-rc1] - 2026-03-16

Added

Changed

  • Bump stackable-operator to 0.108.0, snafu to 0.9, strum to 0.28 (#706, #708).
  • Gracefully shutdown all concurrent tasks by forwarding the SIGTERM signal (#699).
  • Bump testing-tools to 0.3.0-stackable0.0.0-dev (#691).

Removed

  • Remove support for Superset 4.0.2 and 4.1.2 (#695).

Fixed

  • Fix "404 page not found" error for the initial object list (#708).

[25.11.0] - 2025-11-07

[25.11.0-rc1] - 2025-11-06

Added

  • Add end-of-support checker which can be controlled with environment variables and CLI arguments (#670).
    • 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.
  • Helm: Allow Pod priorityClassName to be configured (#664).
  • Add version 4.1.4 (#669).
  • Add prometheus.io/path|port|scheme annotations to metrics service (#671).

Changed

  • Deprecate version 4.0.2 (#666).
  • Bump stackable-operator to 0.100.1 (#674).

Fixed

  • Fix container not starting because Superset was starting too slow and was killed because a failing liveness probe. We now add a proper startup probe, which allows Superset to take longer to start up (#654, #663).

[25.7.0] - 2025-07-23

[25.7.0-rc1] - 2025-07-18

Added

  • Adds new telemetry CLI arguments and environment variables (#623).
    • 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.
  • BREAKING: Added listener support for Superset (#625).
  • Add internal headless service in addition to the metrics service and move listener logic to listener.rs (#644)
  • Add RBAC rule to helm template for automatic cluster domain detection (#646).

Changed

  • BREAKING: Replace stackable-operator initialize_logging with stackable-telemetry Tracing (#610, #617, #623).
    • The console log level was set by SUPERSET_OPERATOR_LOG, and is now set by CONSOLE_LOG_LEVEL.
    • The file log level was set by SUPERSET_OPERATOR_LOG, and is now set by FILE_LOG_LEVEL.
    • The file log directory was set by SUPERSET_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 (#609).
  • test: Bump to Vector 0.46.1 (#628).
  • test: Bump OPA to 1.4.2 (#631).
  • Use versioned common structs (#635).
  • BREAKING: Previously this operator would hardcode the UID and GID of the Pods being created to 1000/0, this has changed now (#637)
    • 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
  • Changed listener class to be role-only (#643).
  • BREAKING: Bump stackable-operator to 0.94.0 and update other dependencies (#646).
    • 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") (#649).

Fixed

  • Use json file extension for log files (#615).
  • Fix a bug where changes to ConfigMaps that are referenced in the SupersetCluster spec didn't trigger a reconciliation (#609).
  • Allow uppercase characters in domain names (#646).

Removed

  • Remove the lastUpdateTime field from the stacklet status (#646).
  • Remove role binding to legacy service accounts (#646).

[25.3.0] - 2025-03-21

Added

  • Run a containerdebug process in the background of each Superset container to collect debugging information (#578).
  • Aggregate emitted Kubernetes events on the CustomResources (#585).
  • Support OPA role mapping as optional custom security manager for Superset (#582).
  • Support for version 4.1.1 (#595).

Changed

  • Bump stackable-operator to 0.87.0 (#602).
  • Default to OCI for image metadata and product image selection (#586).

[24.11.1] - 2025-01-10

Fixed

  • Fix OIDC endpoint construction in case the rootPath does have a trailing slash (#569).
  • 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 (#568).

[24.11.0] - 2024-11-18

Added

  • Allowing arbitrary python code as EXPERIMENTAL_FILE_HEADER and EXPERIMENTAL_FILE_FOOTER in superset_config.py (#530).
  • 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 (#549).

Changed

  • Reduce CRD size from 472KB to 45KB by accepting arbitrary YAML input instead of the underlying schema for the following fields (#528):
    • podOverrides
    • affinity

Fixed

  • Invalid SupersetCluster, DruidConnection or AuthenticationClass objects don't stop the operator from reconciling (#551).

[24.7.0] - 2024-07-24

Added

  • Support for versions 3.1.3 and 4.0.2 (#509).

Changed

  • Bump stackable-operator to 0.70.0, product-config to 0.7.0, and other dependencies (#511).

Fixed

  • Don't print Superset admin credentials during startup (#483).
  • Fix entrypoint to not throw prepare_signal_handlers: command not found in case DB initialization fails (#485).
  • Processing of corrupted log events fixed; If errors occur, the error messages are added to the log event (#502).

Removed

  • Removed unsupported versions 2.1.1, 3.0.1 and 3.0.3 (#509).

[24.3.0] - 2024-03-20

Added

  • Improved CRD docs (#431).
  • Helm: support labels in values.yaml (#448).
  • Add support for OpenID Connect (#423).
  • Support versions 2.1.3, 3.0.3, 3.1.0 (#457).

Changed

  • Raise memory requests and limits for Superset pods to 2Gi (#468).

Fixed

  • BREAKING: Fixed various issues in the CRD structure. clusterConfig.credentialsSecret is now mandatory (#429).

Removed

  • Removed support for version 2.1.0 (#457).

[23.11.0] - 2023-11-24

Added

  • Default stackableVersion to operator version (#390).
  • Support PodDisruptionBudgets (#407).
  • Added support for versions 2.1.1, 3.0.1 (#415).
  • Support graceful shutdown (#422).

Changed

  • vector 0.26.0 -> 0.33.0 (#391, #415).
  • operator-rs 0.44.0 -> 0.55.0 (#390, #407, #415).
  • BREAKING: Removed SupersetDB object, since it created some problems when reinstalling or upgrading a Superset cluster. Instead, the initialization of the database was moved to the startup phase of each Superset pod. To make sure the initialization does not run in parallel, the PodManagementPolicy was set to OrderedReady and liveness/readiness probes were added. The .spec.clusterConfig.loadExamplesOnInit option was removed from the CRD, because loading the examples at every startup caused problems in certain scenarios, e.g. after an upgrade from Superset 1.5.3 to 2.1.0 (#396).

Fixed

  • BREAKING: Rename Service port name from superset to http for consistency reasons. This change should normally not be breaking, as we only change the name, not the port. However, there might be some e.g. Ingresses that rely on the port name and need to be updated (#394).
  • Fixed config override support (#415).

Removed

  • Removed support for versions 1.3.2, 1.4.1, 1.4.2, 1.5.1, 1.5.3, 2.0.1 (#415).

[23.7.0] - 2023-07-14

Added

  • Added support for Superset versions 1.4.2, 1.5.3, 2.0.1 and 2.1.0 (#362).
  • Generate OLM bundle for Release 23.4.0 (#364).
  • Missing CRD defaults for status.conditions field (#367).
  • Set explicit resources on all containers (#371).
  • Support podOverrides (#377).

Changed

  • operator-rs 0.40.2 -> 0.44.0 (#360, #371, #383).
  • Use 0.0.0-dev product images for testing (#361).
  • Use testing-tools 0.2.0 (#361).
  • Added kuttl test suites (#373).
  • [BREAKING] Moved all top level config options to clusterConfig. Authentication is now provided via an array of AuthenticationClasses and additional properties (#379).

Fixed

  • Operator now errors out when credentialsSecret is missing (#375).
  • Increase the size limit of the log volume (#383).

[23.4.0] - 2023-04-17

Added

  • Log aggregation added (#326).
  • Deploy default and support custom affinities (#337).
  • Extend cluster resources for status and cluster operation (paused, stopped) (#348)
  • Cluster status conditions (#349)

Changed

  • [BREAKING]: Support specifying Service type by moving serviceType (which was an experimental feature) to clusterConfig.listenerClass. This enables us to later switch non-breaking to using ListenerClasses for the exposure of Services. This change is breaking, because - for security reasons - we default to the cluster-internal ListenerClass. If you need your cluster to be accessible from outside of Kubernetes you need to set clusterConfig.listenerClass to external-unstable or external-stable (#350).
  • operator-rs 0.31.0 -> 0.35.0 -> 0.40.2 (#322, #326, #352).
  • Bumped stackable image versions to "23.4.0-rc2" (#322, #326).
  • Fragmented SupersetConfig (#323).
  • Restructured documentation (#344).
  • Create ServiceAccount for Superset clusters. Use build_rbac_resources() from operator-rs (#352)

[23.1.0] - 2023-01-23

Changed

  • operator-rs 0.27.1 -> 0.31.0 (#306, #297, #311)
  • Fixed the RoleGroup selector. It was not used before. (#306)
  • Updated stackable image versions (#295)
  • [BREAKING] Use Product image selection instead of version (#304)
    • spec.version has been replaced by spec.image
    • spec.statsdExporterVersion has been removed, the statsd-exporter is now part of the images itself
  • Refactored LDAP authentication handling to use functionality from the LdapAuthenticationProvider (#311)

[0.7.0] - 2022-11-07

Added

  • CPU and memory limits are now configurable (#273).

Changed

  • Don't run init container as root and avoid chmod and chowning (#300).

[0.6.0] - 2022-09-07

Added

  • Add temporary attribute to support using ClusterIP instead of NodePort service type (#266).

Changed

  • Include chart name when installing with a custom release name (#227, #228).
  • Orphaned resources are deleted (#255).
  • operator-rs 0.22.0 -> 0.25.0 (#255).
  • Make webserver timeout configurable. Increase default to 5m (#247).

[0.5.0] - 2022-06-30

Added

  • Configuration option rowLimit added (#173).
  • Configuration and environment overrides enabled (#173).
  • Ability to add MAPBOX_API_KEY from secret added (#178).
  • Update SupersetConfigOptions to include explicit config for MapboxApiKey (#179)
  • Add support for LDAP authentication (#180)
  • [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: 1.4.1 becomes (for example) version: 1.4.1-stackable2.1.0 (#207)
  • Add support for Superset 1.5.1 (#222).

Changed

  • Required product image version changed to 2 (#182).
  • DruidConnection namespace properties are optional now (#187).

Fixed

  • A DruidConnection was not established if the Druid instance was started after the Superset instance, this was fixed (#187).
  • The correct secret key is used when upgrading the Superset database. This issue was introduced in #173 (#190).

[0.4.0] - 2022-04-05

Added

  • Reconciliation errors are now reported as Kubernetes events (#132).
  • Add support for Superset 1.4.1 (#135).
  • Use cli argument watch-namespace / env var WATCH_NAMESPACE to specify a single namespace to watch (#138).

Changed

  • operator-rs 0.9.0 -> 0.13.0 (#132,#138).

[0.3.0] - 2022-02-14

Added

  • Druid DB connections can now be configured in a custom resource (#71).
  • BREAKING: Prometheus metrics enabled (#128); The statsdExporterVersion must be set in the cluster specification.

Changed

  • Shut down gracefully (#70).
  • All dependencies upgraded. The upgrade to operator-rs 0.8.0 does not force the credentials secret to be set anymore in the custom resource but it is still required (#82).
  • operator-rs 0.8.0 → 0.9.0 (#71)

[0.2.0] - 2021-12-17

Changed

  • Migrated to StatefulSet rather than direct Pod management (#45).

[0.1.0] - 2021-12-06

Added

  • Initial implementation of the operator added. An admin user can be created in the Superset database with the Init command which takes the credentials from a secret (#7, #12, #33).