Percona Distribution for MySQL Operator 1.9.0¶

Date

August 9, 2021
Installation

For installation please refer to the documentation page

Release Highlights¶

Starting from this release, the Operator changes its official name to Percona Distribution for MySQL Operator. This new name emphasizes gradual changes which incorporated a collection of Percona’s solutions to run and operate Percona Server for MySQL and Percona XtraDB Cluster, available separately as Percona Distribution for MySQL.
Now you can see HAProxy metrics in your favorite Percona Monitoring and Management (PMM) dashboards automatically.
The cross-site replication feature allows an asynchronous replication between two Percona XtraDB Clusters, including scenarios when one of the clusters is outside of the Kubernetes environment. The feature is intended for the following use cases:
- provide migrations of your Percona XtraDB Cluster to Kubernetes or vice versa,
- migrate regular MySQL database to Percona XtraDB Cluster under the Operator control, or carry on backward migration,
- enable disaster recovery capability for your cluster deployment.

New Features¶

K8SPXC-657: Use Secrets to store custom configuration with sensitive data for Percona XtraDB Cluster, HAProxy, and ProxySQL Pods
K8SPXC-308: Implement Percona XtraDB Cluster asynchronous replication within the Operator
K8SPXC-688: Define environment variables in the Custom Resource to provide containers with additional customizations

Improvements¶

K8SPXC-673: HAProxy Pods now come with Percona Monitoring and Management integration and support
K8SPXC-791: Allow stopping the restart-on-fail loop for Percona XtraDB Cluster and Log Collector Pods without special debug images
K8SPXC-764: Unblock backups even if just a single instance is available by setting the allowUnsafeConfigurations flag to true
K8SPXC-765: Automatically delete custom configuration ConfigMaps if the variable in Custom Resource was unset (Thanks to Oleksandr Levchenkov for contributing)
K8SPXC-734: Simplify manual recovery by automatically getting Percona XtraDB Cluster namespace in the pxc container entrypoint script (Thanks to Michael Lin for contributing)
K8SPXC-656: imagePullPolicy is now set for init container as well to avoid pulling and simplifying deployments in air-gapped environments (Thanks to Herberto Graça for contributing)
K8SPXC-511: Secret object containing system users passwords is now deleted along with the Cluster if delete-pxc-pvc finalizer is enabled (Thanks to Matthias Baur for contributing)
K8SPXC-772: All Service objects now have Percona XtraDB Cluster labels attached to them to enable label selector usage
K8SPXC-731: It is now possible to see the overall progress of the provisioning of Percona XtraDB Cluster resources and dependent components in Custom Resource status
K8SPXC-730: Percona XtraDB Cluster resource statuses in Custom Resource output (e.g. returned by kubectl get pxc command) have been improved and now provide more precise reporting
K8SPXC-697: Add namespace support in the copy-backup script
K8SPXC-321, K8SPXC-556, K8SPXC-568: Restrict the minimal number of ProxySQL and HAProxy Pods and the maximal number of Percona XtraDB Cluster Pods if the unsafe flag is not set
K8SPXC-554: Reduced the number of various etcd and k8s object updates from the Operator to minimize the pressure on the Kubernetes cluster
K8SPXC-421: It is now possible to use X Plugin with Percona XtraDB Cluster Pods

Known Issues and Limitations¶

K8SPXC-835: ProxySQL will fail to start on a Replica Percona XtraDB Cluster for cross-site replication in this release

Bugs Fixed¶

K8SPXC-757: Fixed a bug where manual crash recovery interfered with auto recovery functionality even with the auto_recovery flag set to false
K8SPXC-706: TLS certificates renewal by a cert-manager was failing (Thanks to Jeff Andrews for reporting this issue)
K8SPXC-785: Fixed a bug where backup to S3 was producing false-positive error messages even if backup was successful
K8SPXC-642: Fixed a bug where PodDisruptionBudget was blocking the upgrade of HAProxy (Thanks to Davi S Evangelista for reporting this issue)
K8SPXC-585: Fixed a bug where the Operator got stuck if the wrong user credentials were set in the Secret object (Thanks to Sergiy Prykhodko for reporting this issue)
K8SPXC-756: Fixed a bug where the Operator was scheduling backups even when the cluster was paused (Thanks to Dmytro for reporting this issue)
K8SPXC-813: Fixed a bug where backup restore didn’t return error on incorrect AWS credentials
K8SPXC-805: Fixed a bug that made pxc-backups object deletion hang if the Operator couldn’t list objects from the S3 bucket (e.g. due to wrong S3 credentials)
K8SPXC-787: Fixed the “initializing” status of ready clusters caused by the xtrabackup user password change
K8SPXC-775: Fixed a bug where errors in custom mysqld config settings were not detected by the Operator if the config was modified after the initial cluster was created
K8SPXC-767: Fixed a bug where on-demand backup hung up if created while the cluster was in the “initializing” state
K8SPXC-726: Fixed a bug where the delete-s3-backup finalizer prevented deleting a backup stored on Persistent Volume
K8SPXC-682: Fixed auto-tuning feature setting wrong innodb_buffer_pool_size value in some cases

Contact Us

For free technical help, visit the Percona Community Forum.

To report bugs or submit feature requests, open a JIRA ticket.

For paid support and managed or consulting services , contact Percona Sales.

Last update: 2023-10-12