Releases: k8ssandra/k8ssandra-operator
Releases · k8ssandra/k8ssandra-operator
v1.21.1
v1.21.0
- [CHANGE] #1450 Update datacenter labels to use Kubernetes resource names for CassandraDatacenter, not the cleaned override name. Update to cass-operator 1.23.0
- [CHANGE] #1441 Use k8ssandra-client instead of k8ssandra-tools for CRD upgrades
- [ENHANCEMENT] #1667 Add
skipSchemaMigration
option toK8ssandraCluster.spec.reaper
- [FEATURE] #1034 Add support for priorityClassName
- [BUGFIX] #1454 Do not try to work out backup status if there are no pods
- [BUGFIX] #1383 Do not create MedusaBackup if MedusaBakupJob did not fully succeed
- [BUGFIX] #1460 Fix podName calculations in medusa's hostmap.go to account for unbalanced racks also
- [BUGFIX] #1466 Do not overwrite existing status fields or forget to write the changes. Also, add new ContextName for the Datacenter to know where it used to be.
- [ENHANCEMENT] #1455 Expose configuration of Medusa's gRPC server port
- [BUGFIX] #1471 Use namespaced service name when registering k8ssandra cluster to Reaper
- [BUGFIX] Upgrade cass-operator helm chart to 0.55.0 (1.23.0)
- [CHANGE] Upgrade Reaper to v3.8.0
v1.20.2
v1.20.2 - 2024-10-07
- [DOCS] #1469 Add docs for Reaper's Control Plane deployment mode
- [CHANGE] Bump default Medusa version to 0.22.3
- [BUGFIX] #1409 Vector would crash in the Cassandra log parsing if empty lines were present. Add automated tests for Vector parsing rules.
- [BUGFIX] #1425 prepare-helm-release.sh requires kustomize to be in the path and that makes make manifests fail.
- [BUGFIX] #1422 Changing a DC name should be forbidden
v1.20.1
v1.20.1 - 2024-09-19
- [BUGFIX] Upgrade cass-operator to v1.22.4 to fix security context overwrites
v1.20.0
v1.19.0
v1.18.0
- [CHANGE] Update cassandra-medusa to 0.22.0
- [CHANGE] Update cass-operator to v1.22.0
- [FEATURE] #1310 Enhance the MedusaBackupSchedule API to allow scheduling purge tasks
- [BUGFIX] #1222 Consider DC-level config when validating numToken updates in webhook
- [BUGFIX] #1366 Reaper deployment can't be created on OpenShift due to missing RBAC rule
v1.17.0
- [BUGFIX] #1334 Delete
MedusaBackupJobs
during purge - [BUGFIX] #1336 Propagate labels to the sync
MedusaTask
created after purge - [CHANGE] Update cass-operator to v1.21.0
- [CHANGE] #1313 upgrade controller-runtime to 1.17 series, Go to 1.21.
- [BUGFIX] #1317 Fix issues with caches in cluster scoped deployments where they were continuing to use a multi-namespace scoped cache and not an informer cache.
- [BUGFIX] #1316 Fix interchanged intervals and timeouts in tests.
- [BUGFIX] #1322 Fix bug where server-system-logger customisations from the Containers field would be overwritten when vector was enabled.
- [FEATURE] Add support for HCD 1.0
- [ENHANCEMENT] #1329 Add config emptyDir volume mount on Reaper deployment to allow read only root FS
- [BUGFIX] Fix HCD jvm options generation
- [ENHANCEMENT] #1278 Add toggle to allow the disabling of the Medusa purge CronJob creation
- [BUGFIX] #1299 CronJob for medusa purge not in the correct namespace
- [CHANGE] Upgrade Reaper to v3.6.1