Scylla Operator

Docker Image

docker.io/scylladb/scylla-operator:1.22.0

Changelog

https://github.com/scylladb/scylla-operator/blob/master/CHANGELOG.md#1220

Docker Image

docker.io/scylladb/scylla-operator:1.21.1

Changelog

https://github.com/scylladb/scylla-operator/blob/master/CHANGELOG.md#1211

Docker Image

docker.io/scylladb/scylla-operator:1.20.3

Changelog

https://github.com/scylladb/scylla-operator/blob/master/CHANGELOG.md#1203

Docker Image

docker.io/scylladb/scylla-operator:1.21.0

Changelog

https://github.com/scylladb/scylla-operator/blob/master/CHANGELOG.md#1210

Docker Image

docker.io/scylladb/scylla-operator:1.20.2

Changelog

https://github.com/scylladb/scylla-operator/blob/master/CHANGELOG.md#1202

Docker Image

docker.io/scylladb/scylla-operator:1.19.2

Changelog

https://github.com/scylladb/scylla-operator/blob/master/CHANGELOG.md#1192

Docker Image

docker.io/scylladb/scylla-operator:1.20.1

Changelog

https://github.com/scylladb/scylla-operator/blob/master/CHANGELOG.md#1201

Release notes for 1.20.0

Container images

docker.io/scylladb/scylla-operator:1.20.0

Changes By Kind (since 1.19.1)

Feature

• Add IPv6 support for ScyllaClusters (#2956,@ylebi)

Bug

• Allow for setting additional arguments for Scylla Operator containers through Helm (#3071,@rzetelskik)

Documentation

• Add v1.19 and drop v1.16 from docs (#3057,@rzetelskik)
• Promote v1.19 to latest stable version and deprecate v1.17 in docs (#3061,@rzetelskik)
• Update release schedule (#3062,@rzetelskik)
• Fix address source paths in ServiceLoadBalancerIngress exposure type documentation (#3067,@rzetelskik)
• Document handling multiple clusters in a single Prometheus (#3070,@czeslavo)
• Bring prom-operator back as dependency in installation guide (#3073,@czeslavo)
• Add documentation for feature gates and bootstrap synchronisation (#3084,@rzetelskik)
• Reword platforms to environments (#3112,@czeslavo)
• Document automatic data cleanups (#3133,@czeslavo)
• Update Scylla and Manager Agent versions injected to docs with Renovate (#3163,@czeslavo)
• Fix broken link to OpenShift docs (#3184,@czeslavo)
• Account for possible transient unavailability in Scylla upgrade docs (#3198,@czeslavo)
• Add a custom sphinx extension loading version contexts in build time (#3222,@rzetelskik)
• Unify product naming in docs (#3225,@rzetelskik)
• Update the GA for 1.20 (#3226,@mflendrich)
• Extend support matrix in docs with supported OpenShift version range (#3229,@rzetelskik)
• Make example ScyllaCluster nodes to be considered tunable (#3245,@czeslavo)

Flaky/Failing Test

• List only Scylla nodes pods when getting nodes IPs in E2E (#3114,@czeslavo)
• Fix makeCertificate test race causing panic (#3203,@czeslavo)
• Make timeouts more explicit and improve related error messages in ScyllaDB Manager E2Es (#3252,@rzetelskik)

Other

• Make 'make_latest' parameter in github release API call data a string (#3058,@rzetelskik)
• Bump the gomod group with 10 updates (#3064,@dependabot)
• Downgrade ScyllaDB utils image to 2025.1.9 (#3081,@rzetelskik)
• Remove the workaround for variable substitution in multiversion documentation (#3083,@rzetelskik)
• Sync prom-operator go.mod versions with metadata.yaml (#3088,@czeslavo)
• Drop Dockerfile from OLM bundle directory (#3090,@rzetelskik)
• Add missing fields to OLM bundle patch for fields overwritten during publishing (#3091,@rzetelskik)
• Add repository annotation to OLM bundle CSV metadata (#3092,@rzetelskik)
• Align metadata's min kubernetes version with compatibility matrix (#3093,@rzetelskik)
• Adjust bundle patches to set 'com.redhat.openshift.versions' in metadata annotations instead of CSV manifest (#3094,@rzetelskik)
• Change OLM bundle package name to scylladb-operator (#3095,@rzetelskik)
• Bump actions/checkout from 5 to 6 (#3096,@dependabot)
• Update docs dependencies (#3104,@rzetelskik)
• Bump the gomod group across 1 directory with 20 updates (#3105,@dependabot)
• Use newer AWS AMI AL2023 in the eksctl example (#3106,@abramche)
• Update maintainers in the Helm charts (#3108,@mflendrich)
• Add E2E verifying nodes cleanup after multi-node cluster provisioning (#3111,@czeslavo)
• Install Operator through OLM in non-release deployment scripts for OpenShift (#3116,@rzetelskik)
• Drop operator deployment modifications from CI scripts (#3120,@rzetelskik)
• Add a script for OLM bundle postprocessing (#3121,@rzetelskik)
• Update docs theme to 1.8.10 (#3127,@dgarcia360)
• Bump the gomod group across 1 directory with 12 updates (#3130,@dependabot)
• Extend the OLM bundle postprocessing script with modifications required for a release (#3131,@rzetelskik)
• Update ScyllaDB to 2025.3.5 (#3132,@rzetelskik)
• Set default channel annotation in OLM bundle metadata and make the channel annotation overwritten during publishing (#3137,@rzetelskik)
• Add copilot instructions for commit messages (#3141,@mflendrich)
• Add initial Renovate configuration for updating Scylla (#3142,@czeslavo)
• Add go cache mounts to dockerfile (#3145,@czeslavo)
• Bump the gomod group with 13 updates (#3150,@dependabot)
• Configure Renovate for ScyllaDB Manager and Manager Agent (#3155,@copilot-swe-agent)
• Add a script for building OLM catalog for testing (#3156,@rzetelskik)
• Add initial implementation of envtests (#3157,@czeslavo)
• Update scylladb Docker tag to v2025.4.0 (#3158,@renovate)
• Include the sha256: prefix in Renovate's digest matcher (#3159,@mflendrich)
• Add unit test for matching Manager and Manager Agent tags (#3161,@copilot-swe-agent)
• Replace Dependabot with Renovate for Go modules (#3167,@copilot-swe-agent)
• Group scylla-manager and scylla-manager-agent Renovate updates (#3172,[@copilot-swe-agent]

…(truncated)

Release notes for 1.19.1

Container images

docker.io/scylladb/scylla-operator:1.19.1

Changes By Kind (since 1.19.0)

Documentation

• Bring prom-operator back as dependency in installation guide (#3076,@czeslavo)
• Update release schedule (#3086,@rzetelskik)
• Add documentation for feature gates and bootstrap synchronisation (#3087,@rzetelskik)
• Fix broken link to OpenShift docs (#3235,@rzetelskik)

Flaky/Failing Test

• List only Scylla nodes pods when getting nodes IPs in E2E (#3135,@czeslavo)

Other

• Update go to 1.25 and adapt to internal crypto changes (#3235,@rzetelskik)
• Align myst_substitutions with master source (#3099,@rzetelskik)
• Use newer AWS AMI AL2023 in the eksctl example (#3110,@abramche)
• Update dependencies (#3235,@rzetelskik)

Release notes for 1.19.0

Container images

docker.io/scylladb/scylla-operator:1.19.0

Changes By Kind (since 1.18.1)

API Change

• Add support for RetryWait parameter in ScyllaDB Manager tasks' integration (#2833,@rzetelskik)
• Deprecate exposeOptions in ScyllaDBMonitoring (#3014,@czeslavo)
• Return warnings from admission webhook on deprecated ScyllaDBMonitoring fields (#3038,@czeslavo)

Feature

• Extend admission review with warning messages (#2893,@rzetelskik)
• Extend node tuning with kernel parameter configuration (sysctls) (#2901,@rzetelskik)
• Add support for external Prometheus mode to ScyllaDBMonitoring (#2943,@czeslavo)
• Ensure all nodes in the cluster see all as UP as a prerequisite for adding new nodes (#2992,@rzetelskik)

Bug

• Fix optional description formatting in ScyllaDBManager task completion verification (#2815,@rzetelskik)
• Don't provision 2025.3 and 2025.3.0 Grafana platform dashboards (#2823,@czeslavo)
• Fix Grafana container image reference to use a multi-arch index (#2829,@rzetelskik)
• Do not create perftune jobs when NodeConfig has optimizations disabled (#2875,@rzetelskik)
• Fix potential data races in RenderTemplate function (#2954,@czeslavo)
• Don't emit FilesystemCreated when device already formatted (#2989,@czeslavo)
• Fix condition reason overflow omission indicator to match upstream regex (#2998,@czeslavo)
• Always set rack name in ScyllaDBDatacenter rack status (#3000,@rzetelskik)
• Allow for setting additional arguments for Scylla Operator containers through Helm (#3072,@rzetelskik)

Documentation

• Update docs theme 1.8.7 (#2747,@dgarcia360)
• Add v1.18 to docs (#2808,@czeslavo)
• Fix variable substitutions for multiversion documentation (#2824,@zimnx)
• Add a dedicated upgrade guide (#2837,@czeslavo)
• Fix broken links in docs (#2838,@czeslavo)
• Add the 1.18 release and the 1.19 plans (#2872,@czeslavo)
• Add ScyllaDB version upgrade guide to documentation (#2880,@rzetelskik)
• Fix docs issues found in v1.18 testing (#2886,@czeslavo)
• Add Management section in docs and restructure operator upgrade guide (#2888,@rzetelskik)
• Update example operator tag in GitOps installation guide to 1.18.0 release (#2891,@abramche)
• Add GKE (Ubuntu) >= 1.32 to unsupported platforms (#2892,@czeslavo)
• Handle missing xfsprogs on GKE >= 1.32 (#2897,@czeslavo)
• Reorganize K8s prerequisites pages (#2898,@czeslavo)
• Update supported Kubernetes versions (#2902,@czeslavo)
• Add ScyllaDBMonitoring external Prometheus support enhancement proposal (#2909,@czeslavo)
• Add API conventions document (#2919,@czeslavo)
• Add operator-sdk dependency to contributing guide (#2925,@czeslavo)
• Add sysctl configuration guide to documentation (#2939,@rzetelskik)
• Remove number-lines directive from GKE systemconfig.yaml (#2950,@abramche)
• Add an enhancement proposal for a failed replace recovery guide (#2955,@mflendrich)
• Remove number-lines directive from all help files (#2957,@abramche)
• Update documentation theme to 1.8.8 (#2960,@dgarcia360)
• Extend tagged union convention with options validation (#2961,@czeslavo)
• Link to up-to-date release procedures in RELEASE.md (#2971,@czeslavo)
• Document namespace requirements and best practices (#2976,@mflendrich)
• Unify NodeConfig resource name in docs and examples (#2978,@rzetelskik)
• Rework and split ScyllaDBMonitoring docs into granular articles (#2993,@czeslavo)
• Extend monitoring docs with external Prometheus mode (#3005,@czeslavo)
• Update the 1.19 release date (#3016,@mflendrich)
• Fix monitoring overview redirect (#3025,@czeslavo)
• Refresh project readme to rely more on docs (#3026,@czeslavo)
• Fix Prometheus Operator installation instructions in docs (#3029,@rzetelskik)
• Fix HAProxy Ingress installation instructions in docs (#3030,@rzetelskik)
• Update ScyllaDB version in docs to 2025.3.2 (#3033,@rzetelskik)
• Explicitly mention dependency updates and clarify versionn-specific steps in the upgrade guide (#3040,@rzetelskik)
• Remove line numbers from example manifest in sysctl guide (#3041,@rzetelskik)
• Update the 1.19 release date (#3043,@mflendrich)
• Adjusts monitoring diagram to external Prometheus (#3047,@czeslavo)
• Remove reference to ScyllaDBMonitorings from Resources box in docs overview (#3049,@rzetelskik)
• Fix address source paths in ServiceLoadBalancerIngress exposure type documentation (#3068,@rzetelskik)
• Document handling multiple clusters in a single Prometheus (#3074,@czeslavo)

Flaky/Failing Test

• Explicitly set and unify the number of retries and backoff duration in ScyllaDB Manager tasks in E2E tests (#2834,@rzetelskik)
• Extend timeout in DynamicSharedInformerFactory UT (#2835,@czeslavo)
• Await schema agreement after key

…(truncated)