Percona Kubernetes Operator for Percona XtraDB Cluster 1.7.0

• Date

  February 2, 2021

• Installation

  Installing Percona Kubernetes Operator for Percona XtraDB Cluster

New Features

• K8SPXC-530: Add support for point-in-time recovery

• K8SPXC-564: PXC cluster will now recover automatically from a full crash when Pods are stuck in CrashLoopBackOff status

• K8SPXC-497: Official support for Percona Monitoring and Management (PMM) v.2

NOTE: Monitoring with PMM v.1 configured according to the unofficial instruction will not work after the upgrade. Please switch to PMM v.2.

Improvements

• K8SPXC-485: Percona XtraDB Cluster Pod logs are now stored on Persistent Volumes. Users can debug the issues even after the Pod restart

• K8SPXC-389: User can now change ServiceType for HAProxy replicas Kubernetes service

• K8SPXC-546: Reduce the number of ConfigMap object updates from the Operator to improve performance of the Kubernetes cluster

• K8SPXC-553: Change default configuration of ProxySQL to WRITERS_ARE_READERS=yes so Percona XtraDB Cluster continues operating with a single node left

• K8SPXC-512: User can now limit cluster-wide Operator access to specific namespaces (Thanks to user mgar for contribution)

• K8SPXC-490: Improve error message when not enough memory is set for auto-tuning

• K8SPXC-312: Add schema validation for Custom Resource. Now cr.yaml is validated by a WebHook for syntax typos before being applied. It works only in cluster-wide mode due to access restrictions

• K8SPXC-510: Percona XtraDB Cluster operator can now be deployed through RedHat Marketplace

• K8SPXC-543: Check HAProxy custom configuration for syntax errors before applying it to avoid Pod getting stuck in CrashLoopBackOff status (Thanks to user pservit for reporting this issue)

Bugs Fixed

• K8SPXC-544: Add a liveness probe for HAProxy so it is not stuck and automatically restarted when crashed (Thanks to user pservit for reporting this issue)

• K8SPXC-500: Fix a bug that prevented creating a backup in cluster-wide mode if default cr.yaml is used (Thanks to user michael.lin1 for reporting this issue)

• K8SPXC-491: Fix a bug due to which compressed backups didn’t work with the Operator (Thanks to user dejw for reporting this issue)

• K8SPXC-570: Fix a bug causing backups to fail with some S3-compatible storages (Thanks to user dimitrij for reporting this issue)

• K8SPXC-517: Fix a bug causing Operator crash if Custom Resource backup section is missing (Thanks to user deamonmv for reporting this issue)

• K8SPXC-253: Fix a bug preventing rolling out Custom Resource changes (Thanks to user bitsbeats for reporting this issue)

• K8SPXC-552: Fix a bug when HAProxy secrets cannot be updated by the user

• K8SPXC-551: Fix a bug due to which cluster was not initialized when the password had an end of line symbol in secret.yaml

• K8SPXC-526: Fix a bug due to which not all clusters managed by the Operator were upgraded by the automatic update

• K8SPXC-523: Fix a bug putting cluster into unhealthy status after the clustercheck secret changed

• K8SPXC-521: Fix automatic upgrade job repeatedly looking for an already removed cluster

• K8SPXC-520: Fix Smart update in cluster-wide mode adding version service check job repeatedly instead of doing it only once

• K8SPXC-463: Fix a bug due to which wsrep_recovery log was unavailable after the Pod restart

• K8SPXC-424: Fix a bug due to which HAProxy health-check spammed in logs, making them hardly unreadable

• K8SPXC-379: Fix a bug due to which the Operator user credentials were not added into internal secrets when upgrading from 1.4.0 (Thanks to user pservit for reporting this issue)

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: 2023-10-02