github.com/bitpoke/mysql-operator

⚠️ released on 2021-12-28.

Fixed

• orchestrator.secretName is ignored in helm charts

⚠️ This release drops support for Helm 2 and adopts Kubernetes Recommended Labels for the helm deployment.

Please check upgrade instructions and blog article before upgrading.

Fixed

• Fix the app version in the published Helm charts

⚠️ This release drops support for Helm 2 and adopts Kubernetes Recommended Labels for the helm deployment.

Please check upgrade instructions and blog article before upgrading.

Added

• If you want to save MySQL backup to AWS S3, AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY were the only options, but now you can use AWS_SESSION_TOKEN or AWS_ROLE_ARN and AWS_WEB_IDENTITY_TOKEN_FILE
• Add orchestrator.persistence.selector.matchLabels and orchestrator.persistence.annotations for
  persistence depolyment with constraints
• Add orchestrator.persistence.fsGroupWorkaroundEnabled for persistent volume
  provisioners wich don't support fsGroup in security context (fixes #615)
• Add appSecretLabels, appSecretAnnotations, backupSecretLabels, backupSecretAnnotations to provide
  custom labels and annotations to created app and backup secrets
• Update rclone to v1.57.0
• For s3, enable the no_check_bucket option for rclone
• Allow setting metrics and health checking listening addresses

Changed

• Allow setting pod security context when deploying with Helm
• Use distroless as base image for orchestrator container
• Use networking.k8s.io/v1 instead of extensions/v1beta1 for ingress
• Use Service.spec.publishNotReadyAddresses instead of service.alpha.kubernetes.io/tolerate-unready-endpoints
• Use git-semver for setting versions
• Graceful shutdown is enabled by default

Removed

• Removed support for Helm 2

Fixed

• Make sure orchestrator can find its templates #741
• Workaround CRD generation with preserveUnknownFields=false kubernetes-sigs/controller-tools#476

Fixed

• Workaround CRD generation with preserveUnknownFields=false kubernetes-sigs/controller-tools#476

Added

• If you want to save mysql backup to AWS S3, AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY were the only options, but now you can use AWS_SESSION_TOKEN or AWS_ROLE_ARN and AWS_WEB_IDENTITY_TOKEN_FILE

Changed

• Update rclone to v1.57.0

Fixed

• Make sure orchestrator can find it's templates #741

Added

• Add orchestrator.persistence.fsGroupWorkaroundEnabled for persistent volume
  provisioners wich don't support fsGroup in security context (fixes #615)

Changed

• Allow setting pod security context when deploying with Helm
• Use distroless as base image for orchestrator container

⚠️ Helm charts and docker images location changed

• Helm charts repository changed to https://helm-charts.bitpoke.io
• Docker images are now published to https://hub.docker.com/u/bitpoke

Added

• Add image and mysqlVersion options to MysqlCluster chart. This bumps the chart version to 0.3.1
• Add backupAffinity, backupNodeSelector, backupPriorityClassName, backupTolerations
  to .Spec.PodSpec to allow specifying custom scheduling constraints for backup jobs.
• Add the ability to set the imagePullSecrets for the operator statefulset.
• Add Google Drive via service account as backup option.
• Add initBucketURL and initBucketSecretName options to MysqlCluster chart. This bumps the chart version to 0.3.0
• Add an example of how initContainers can be used to fix hostPath permissions.
• Add a lifecycle preStop hook for the mysql container. Before killing the master MySQL process,
  it triggers a graceful-master-takeover-auto command in Orchestrator. This is disabled by
  default, to enable it set gracefulShutdown.enabled=true in chart values or set the controller
  command argument failover-before-shutdown to true.
• Add mysqlLifecycle to .Spec.PodSpec to allow overriding the default lifecycle hook for the
  mysql container.
• Add backupCompressCommand and backupDecompressCommand to allow using
  different compressors/decompressors when backing up or restoring.
• Add support for MySQL version 8.0
• Add go modules cache
• Support cron timezone

Changed

• Only add binlog-space-limit for percona image
• Make user-defined InitContainer take the precedence
• Set timeout of 15s on connection between the operator and Orchestrator
• Bump controller-util dependency to 0.3.0 which fixes some updates on pod spec.
• Removed NO_AUTO_VALUE_ON_ZERO from sql-mode to be inline with MySQL default value
• Remove use go module instead of dep
• Update k8s client to v0.21.2
• Update kubebuilder (to v2.3.1) along with controller-runtime (to v0.9.2) and controller-gen
• Update rclone to v1.53.1
• Update Orchestrator version from v3.1.2 to v3.2.3
• Set default MySQL server version to 5.7.31
• Generate CRDs with controller-gen v0.5.0
• Update getOrdinalFromHostname and IsFirstPodInSet
• Use klog for logging
• Use https://github.com/bitpoke/build for building the project

Fixed

• Fix insufficient permissions during startup
• Fix the xtrabackup --tables-exclude cannot take effect
• Fix the pod unable to connect Orchestrator
• Fix pod labels diff of map
• Fixed backup cleanup job bug (#577)
• Fix Kubebuilder path in Makefile.
• Fix #632 lifeCycle preStop script is not copied to given directory.
• Fix #637 mysqlbackup status is not updated correctly.
• Fix #647 custom conf can't overwrite the default conf
• Fix #627 let Orchestrator do the failover
• Fix #694 the error caused by backing up without waiting for master to be created.

🎉 With this release let's welcome @cndoit18 as a project contributor and maintainer!

Added

• Added a ServiceMonitor to the operator chart that scrapes all MySQL instances.
• Added a test suite for RunCloneCommand logic, along with a mock backup server.
• Added checks for service availability when cloning.
• Added "fail fast" logic when unexpected errors occur during cloning/download.
• Added dataDir cleanup code so that interrupted cloning does not leave dataDir in an inconsistent state.
• Added e2e test demonstrating cloning failure when PVC is removed and pod recreated.
• Add MetricsExporterExtraArgs field on MySQLCluster resource that allows to specify command line arguments to pass to MySQL metrics exporter.
• Allow using custom secret for app credentials in mysql-cluster chart.
• Add XbstreamExtraArgs field on MySQLCluster resource that allows to specify extra command line arguments to xbstream.
• Add XtrabackupExtraArgs, XtrabackupPrepareExtraArgs, XtrabackupTargetDir to parametrize xtrabackup.
• Add RcloneExtraArgs to parametrize rclone command.
• Add InitFileExtraSQL to insert custom init SQL queries that will be run at MySQL initialization.
• Add Volumes and VolumesMounts in .Spec.PodSpec to allow the user specifying custom volume mounts.
• Add InitContainers and Containers in .Spec.PodSpec to allow the user specifying custom containers.
• Add MetricsExporterResources and MySQLOperatorSidecarResrouces in .Spec.PodSpec to allow the user specifying resources for thos sidecars containers.
• Add command line flag to configure number of workers for orchestrator controller.
• Add livenessProbe to controller

Changed

• #422 adds the SidecarServerPort to the MasterService and introduces one new service, HealthyReplicasService, so that we can try to clone from replicas first, then fall back to master.
• Changed the connect timeout from the default of 30s to 5s so that an empty k8s service will not cause cloning attempts to hang unnecessarily for 30s.
• Update documentation structure and formatting.
• Update Orchestrator version to v3.1.4
• Update orchestrator base image to alpine:3.11.
• Update FailoverInProgress condition to false when both Replicas and ReadyNodes are 0.
• Fall back to current master, not pod 0, when no healthy replicas found for backup candidate.
• Change the mysql-operator chart to be helm v3 compatible while keeping backward compatibility.
• Change logging: change cluster logging field to key; normalize logging and more details; output Stackdrive compatible format.
• Decrease MASTER_CONNECT_RETRY interval from 10 to 1 second.
• Deprecate TmpfsSize because can be handled using ExtraVolumes and ExtraMySQLVolumesMounts.
• Update cron documentation
• Set InnoDB buffer parameter: innodb_buffer_pool_instances to min(resources.limit.cpu, floor(innodb_buffer_pool_size/1G)) (see #502)
• Change default resource requests/limits for sidecar container: requested=10m/32Mi, limit=the same as .spec.podSpec.resources.limit
• Change default resource requests/limits for exporter container: requested=10m/32Mi, limit=100m/128Mi
• Change default resource requests/limits for heartbeat container: requested=10m/32Mi, limit=100m/64Mi
• If extra_port is defined in the cluster spec, metrics exporter will use it to connect to MySQL providing that extra_max_connections is larger than the default 1. If MySQL server runs out of available connections, using extra_port allows the exporter to continue collecting MySQL metrics.
• Change the default number of workers for orchestrator controller from 1 to 10.

Removed

Fixed

• Update and fix e2e tests
• Fix double date string in backup path
• Fix double date string in bakup path
• Copy the nodeSelector as-is in the statefulset (fixes #454)
• Fix flakines in ReadOnly cluster condition (fixes #434)
• Fix rounding in computing innodb-buffer-pool-size (fixes #501)
• rclone extra arguments are now properly passed to the backup job.

Fixed

• Fixed TmpfsSize type in CRD

Added

• Add support for mounting a tmpfs into /tmp. Add TmpfsSize field on the cluster

Added

• Update mysql-cluster chart to support setting backupScheduleJobsHistoryLimit

Changed

• Increase size of value column in sys.operator table (see #447) (fixes #446)
• Determine master logic: prevent to follow infinit loops

Fixed

• Use custom server offset (MyServerIDOffset) when deciding to clone from bucket

Changed

• Update Orchestrator version to v3.1.2
• Update Go lang to 1.13.4

Fixed

• Don't require backupSecretName if not set. Fixes delete on remove when using workload identity

Added

• Update mysql-cluster chart to support custom server id offset

Changed

• During failover don't interfere with readdable/wriatable master (fixes #411)
• Imorove logging

Added

• Allow server id offset to be set via MySQLCluster resource by adding MyServerIDOffset field

Changed

• Make app credentials optional for mysql-cluster chart
• Don't create user with empty password (fixes #385)

Fixed

• Fix wrong enviroment for mysql container
• Fix pt-kill client configure file prefix

Changed

• Update docs structure on presslabs.com/docs/mysql-operator/
• Set limit on mysql mysql-init-only container the same as on the mysql container (fixes #371)
• Don't limit memory on sidecars containers

Fixed

• Upgrade path from version v0.2.x (#369)
• Changes the default leader election id from controller-leader-election-helper to mysql-operator-leader-election (#368)

Added

• add a new cluster condition FailoverInProgress that marks the cluster during a failover
• set orchestrator related events on the cluster: OrcFailureDetection, OrcPostUnsuccessfulFailover, OrcPostMasterFailover, OrcPostIntermediateMasterFailover
• new command line flag --mysql-versions-to-image to allow user to specify image for a mysql version
• add print column in mysqlcluster CRD: Ready, Replicas, Age
• allow specifying more PodSpec on MySQL cluster (9b6b46f)
• add a node controller for MySQL configuration (1950812)
• add "standard" labels on services created by the operator (#299)

Changed

• merge Orchestrator chart with the mysql-operator chart. Now instead of deployment it uses a statefulset
• nodes are removed from cluster status at scale down
• use init container for MySQL initialization (#342)
• enhance Backup Job Pod for Workload Identity (#366)
• refactor of how the information flow works: from k8s -> Operator -> Orchestrator; the sidecar container does not connect to Orchestrator anymore.
• rename of initBackupURI to initBackupURL (a3c6556)
• use of Percona CentOS based images (#254)
• don't run as root user in containers (#291)
• rename orchestrator finalizer (to block cluster deletion while it's registered into Orchestrator) from OrchestratorFinalizer to mysql.presslabs.org/registered-in-orchestrator (bfe4646)
• improvement of getBackupCandidate function (9ce4e68)
• configure MySQL using init-file (beb41ce, e5823cb)
• rename AWS_SECRET_KEY field from bucket/backup secret to AWS_SECRET_ACCESS_KEY (#301)
• use an internal status table to store the MySQL status (181909f)
• use a common headless service for all MySQL nodes to reduce the host-name length (#246)

Removed

• cleanup of deprecated fields: bucketURI and top-level MySQLCluster.spec.volumeSpec PVC specification (d909ab9, df80b28)

Fixed

• fix readiness probe for MySQL container
• fix remote storage delete bug
• fix mysqlbackup deletion failure when cluster is not found
• fix #350 where no backups were created when one fails
• fix #107, the orchestrator RAFT issue