🚀 Release 1.8.0-rc1 (2024-01-29)

🗒 NOTES

StackGres 1.8.0-rc1 is out! 🎊 🍾

We introduced 2 new sharding technologies to our SGShardedCluster:

• Apache ShardingSphere (requires installation of the ShardingSphere operator)
• DDP, an in-house sharding technology based on Postgres partitioning functionality and Postgres Foreign Data Wrapper extension.

This release include also some improvements and bug fixes for SGBackup.

So, what you are waiting for to try this release and have a look to the future of StackGres!

✨ NEW FEATURES AND CHANGES

• Added support for DDP as a sharding technology based on FDW
• Added support for ShardingSphere as a sharding technology
• Added securityUpgrade to SGShardedDbOps
• Allow to set secretref as source for the admin credentials
• Add can-i for SGConfig in REST API
• Validate SGInstanceProfile resources are not negative
• Changed default initdb auth host to scram-sha-256 and configurable via password_encryption
• Added timeouts to backups and improved logging and errors

Web Console

• Support spec.profile for sgdistributedlogs
• Support spec.profile for sgshardedclusters
• Support K8s snapshots for SGCluster backups configuration
• Support K8s snapshots for SGShardedCluster backups configuration
• Support volumeSnapshot spec for SGBackups
• Graphic PITR selection feature on sgclusters creation form

🐛 FIXES

• REST API ServiceAccount can not be deleted since owned by OLM
• Workaround for operator bundle upgrade where service account stackgres-restapi is being deleted somehow
• Skip SGConfig reconciliation when lock is updated
• Operator reconcile Web Console nginx ConfigMap in an endless loop when webCertName is set
• SGBackup is marked as Completed when the backupInformation is not set
• Relocate binaries fail when cp is interrupted while copying files
• Postgres exporter uses hardcoded username
• Connections panel GAUGE bar not working as expected

Web Console

• Pod selector not working on cluster monitoring tab
• Replace pod ip with pod's name on sgcluster monitoring url
• Include dashboard selection dropdown on sgshardecluster monitoring tab
• Adjust sidebar namespace selector behavior
• Make sure only one sidebar crd submenu is open at a time
• Add default state to disableclusterpodantiaffinity field on distributed logs form
• Add default state to disableclusterpodantiaffinity field on sharded cluster form
• Show sgbackup tablespaceMap info only when not null
• Move user suppliedd pods sidecars to sidecars steps on sharded cluster form
• Make sharded cluster name on breadcrumbs clickable
• Collapse repeaters on cluster form
• Add classnames to support tests on cluster form
• Collapse repeaters on shardedcluster form
• Hide empty specs on sharded cluster summary

🚧 KNOWN ISSUES

• Major version upgrade fails if some extensions version are not available for the target Postgres version (#1368)
• Backups may be restored with inconsistencies when performed with a Postgres instance running on a different architecture (#1539)

🆙 UPGRADE

To upgrade from a previous installation of the StackGres operator's helm chart you will have to upgrade the helm chart release. For more detailed information please refer to our documentation.

To upgrade StackGres operator's (upgrade only works starting from 1.1 version or above) helm chart issue the following commands (replace namespace and release name if you used something different):

helm upgrade -n "stackgres" "stackgres-operator" https://stackgres.io/downloads/stackgres-k8s/stackgres/1.8.0-rc1/helm/stackgres-operator.tgz

IMPORTANT: This release is incompatible with previous alpha or beta versions. Upgrading from those versions will require uninstalling completely StackGres including all clusters and StackGres CRDs (those in stackgres.io group) first.

Thank you for all the issues created, ideas, and code contributions by the StackGres Community!

🔀 FULL LIST OF COMMITS