v0.8.1

Bug Fixes

• Miscellaneous bugfixes and hardening for the "managed scaling" feature adding in v0.8.0.
• Init-containers now avoid writing to "/tmp" and other root FS locations, to better support "read-only" root filesystems
• SolrPrometheusExporter no longer fails liveness probes when the SolrCloud is too large

v0.8.0

Upgrade Notes & Compatibility Warnings

• The minimum supported version of Solr has been set to Solr 8.11
• The minimum Kubernetes version supported is now v1.22
• Please refer to the documentation for a complete list of upgrade warnings

Improvements

• Solr 8.11 features are now supported by default, such as maxBooleanClauses, metrics disabling, health endpoint for readinessCheck
• Rolling Updates for SolrClouds using ephemeral storage are now safer and replicas are balanced at the end of the operation to ensure optimal resource utilization.
  • Replica balancing is only supported when Solr 9.3+ is used.

New Features

• Managed scale up and scale down are now supported for SolrClouds.
  • By default, when scaling down a SolrCloud, replicas will be migrated off Pods before they are deleted.
  • By default, when scaling up a SolrCloud, replicas will be balanced across all Pods after the SolrCloud has been scaled up. (Only supported for Solr 9.3+)
• Keystore/Truststore passwords can be explicitly set in the SolrCloud CRD for mountedDir SSL. This enables the use of the CertManager CSI Driver with Solr.

Bug Fixes

• SSL bugs with Solr 9 have been fixed, and v0.8.0 will successfully support SSL for Solr 8.11 and 9.4+

v0.7.1

Upgrade Notes & Compatibility Warnings

• Please refer to the documentation for a complete list of upgrade warnings

Bug Fixes

• Upgraded the Kubernetes API clients to fix crash when running with Kubernetes 1.27. #583

v0.7.0

Upgrade Notes & Compatibility Warnings

• Please refer to the documentation for a complete list of upgrade warnings
• The minimum supported Kubernetes version is now v1.21.
• The required Zookeeper Operator version is now v0.2.15. #549
• Options deprecated in v0.6.0 have been removed. #553

New Features

• Clusterwide PodDisruptionBudgets are now created for SolrClouds by default. #471 and #538
• SolrBackups will now backup all collections if no collections are provided. #515

Improvements

• During a rolling restart/update, SolrCloud pods are now taken out of the common service endpoint before they are decommissioned. This should improve request reliability during rolling restarts. #529
• Many missing Provided Zookeeper options have been added. #477, #534 and #557
• Using volume backupRepositories now initialize faster on pod creation and work with custom user/fsGroup. #537
• Default SolrCloud and SolrPrometheusExporter Pod probes have been improved. #510
• The Solr Operator now patches Solr resource statuses. This will reduce needless error logging. #544
• Custom annotations can be added to serviceAccounts created by the helm charts. #480
• The SolrPrometheusExporter service should now work with non-operator Prometheus setups. #483

Bug Fixes

• Non-recurring backups now work as-expected. #509
• Custom persistent data volume names no longer cause the Solr Operator to crash. #479
• Volume backupRepositories now work with custom runAsUser and fsGroup options. #519

v0.6.0

Upgrade Notes & Compatibility Warnings

• Please refer to the ArtifactHub changelog for a complete list of changes

Bug Fixes

• Users are now able to specify resource constraints for all init-containers.
• The default Solr image version used in newly created solrcloud's has been updated from 8.9 to 8.11
• solrcloud now has a 'solrZkOpts' field, allowing users to specify any system properties needed to connect to their ZooKeeper cluster.
• Users running in GKE can now omit the 'gcsCredentialSecret' field from the GCS backup repositories, and fallback on "Workload Identity" and other "credential-less" authentication mechanisms offered by Google Cloud.

v0.5.1

Upgrade Notes & Compatibility Warnings

• Please refer to the documentation for a complete list of upgrade warnings
• The project now uses GoLang 1.17

Bug Fixes

• Fix security vulnerabilities introduced by dependencies
• Ability to run sidecars of the Solr Operator that use the root user, when installing via the Helm Chart
• The zk.external option in the solr helm chart is now functional
• Resource checking for pods across all generated resources has been fixed, will not overwrite equivalent resources with different units
  • Necessary step for GKE Autopilot and the Vertical Pod Autoscaler
• Support ArgoCD for the CRDs
• Fix bugs in documentation

v0.5.0

Upgrade Notes & Compatibility Warnings

• Please refer to the documentation for a complete list of upgrade warnings
• The minimum supported Kubernetes version is now v1.19
• Remove Persistence option for SolrBackup. by @HoustonPutman in #357

New Features

• Support custom Lifecycle for Solr & PrometheusExporter containers by @vladiceanu in #324
• Allow users to take GCS-based backups by @gerlowskija in #302
• Add support for Solr Modules and additional libs by @HoustonPutman in #332
• Add support for S3Repositories by @HoustonPutman in #345
• Add ability to set pod topologySpreadConstraints by @HoustonPutman in #350
• Support to bootstrap from a user-supplied security.json and minor refactorings to pass auth headers through a Context by @thelabdude in #356
• Enable Solr Operator metrics when using the helm chart. by @HoustonPutman in #360
• Ability to set ingressClassName for v1 Ingress. by @HoustonPutman in #363
• Add support for more ZK Pod options. by @HoustonPutman in #361
• Specify individual backupRepo availability in SolrCloud Status by @HoustonPutman in #358
• Add ability to schedule recurring incremental backups by @HoustonPutman in #359

Improvements

• Upgrade to use Kubebuilder 3 by @HoustonPutman in #321
• Allow for managed restarts across multiple SolrCloud resources. by @HoustonPutman in #349
• Enable leader election for the operator by default. by @HoustonPutman in #367
• Add example on how to trigger manual restart to docs by @nosvalds in #354
• Refactor security related code into a separate solr_security_util.go vs. sprinkled throughout the codebase. by @thelabdude in #334
• Start documenting Solr features supported in future versions. by @HoustonPutman in #342
• Clear up difference between SolrCloud CRD and Solr Helm chart options by @HoustonPutman in #362
• Better messaging around supported Solr images. by @HoustonPutman in #373

Bug Fixes

• Fix imagePullSecrets bug when updating Zookeeper Clusters by @HoustonPutman in #344
• Properly resolve the name of the basic auth secret for a SolrCloud when calling the backup API by @thelabdude in #370

v0.4.0

Upgrade Notes & Compatibility Warnings

• Please refer to the documentation for a complete list of upgrade warnings
• The Zookeeper Operator dependency has been upgraded to v0.2.12 (#271)
• Removed deprecated Solr Operator Helm chart option useZkOperator, use zookeeper-operator.use instead (#286)

New Features

• A helm chart to template and deploy SolrCloud resources (#112)
• Ability to schedule automatic restarts for SolrClouds (#281)
• Ability to use HostPath volumes for ephemeral Solr storage (#266)
• Customize serviceAccountName for SolrCloud and SolrPrometheusExporter (#264)
• Introduced ephemeral option for Zookeeper storage (#259)
• Ability to customize probes for PrometheusExporter (#282)
• Add a mountedServerTLSDir config option to support a unique certificate per pod mounted dynamically by an external agent or CSI driver (#291)
• Ability to terminate TLS at Ingress for SolrCloud (#268)
• Ability to schedule automatic restarts for SolrPrometheusExporters (#310)
• Ability to specify ZK Config properties for provided Zookeeper Clusters (#290)
• Option to watch for updates to the mTLS client certificate used by the operator to call Solr pods (#317)
• Configuration options to support an additional client TLS cert in addition to the server certificate (#300)

Improvements

• Default Solr Version upgraded to 8.9, does not affect existing clouds (#285)
• Changed Solr Operator base Docker image to reduce vulnerabilities (#294)
• Grant access to the /admin/zookeeper/status path to the k8s role in the initial security.json (#289)

Bug Fixes

• Remove users role from the all permission in the initial security.json (#274)

v0.3.0

Upgrade Notes & Compatibility Warnings

• Please refer to the documentation for a complete list of upgrade warnings.
• The Solr Operator is now an Apache project managed by the Apache Solr PMC.
  • The CRDs are now under the solr.apache.org group instead of solr.bloomberg.com.
• The Zookeeper Operator is now up-to-date with the most recent release, v0.2.9 (#231)

New Features

• The Solr Operator now fully supports running Solr in a secure and locked down way
  • Basic Auth support is now built in when requested in the SolrCloud CRD (#218)
  • Solr can be run with TLS (optional mTLS, if provided to the operator) (#209, #255)
• Support custom log4j2 config from user-provided ConfigMap (#193)
• Override Prometheus exporter config XML via a user-supplied ConfigMap (#189)
• The Zookeeper Operator can be installed as a helm chart dependency, with the Solr Operator (#231)
• Added support for the Kubernetes Horizontal Pod Autoscaler (HPA) (#254)

Improvements

• Allow the definition of multiple imagePullSecrets (#216)
• More helm chart options are provided to customize running the Solr Operator (#250)
• Solr pod shutdown is more graceful, has better coordination between Kubernetes and Solr (#226)

Bug Fixes

• Fixed ignored options in podOptions probes (#208)

Miscellaneous

• Upgrade Go to 1.16 (#239)
• Use k8s.io/api/networking/v1beta1 Ingress. (#217)

v0.2.8 - Upgrade Bugfix

This will be the final v0.2.* Solr Operator release, and the final bloomberg/solr-operator release.
All future releases can be found at apache/solr-operator.

Bug Fixes

• Fixing upgrade issue with new Status fields (#170)