Releases: aiven/aiven-operator
Releases · aiven/aiven-operator
v0.34.0
- Change service version fields: removed enum validation from
userConfig
(affectsAlloyDBOmni.userConfig.alloydbomni_version,
Cassandra.userConfig.cassandra_version,Flink.userConfig.flink_version,
MySQL.userConfig.mysql_version,OpenSearch.userConfig.opensearch_version,
Redis.userConfig.redis_version,PostgreSQL.userConfig.pg_version,Kafka.userConfig.kafka_version) - Change
ServiceIntegration: operator now adopts existing integrations instead of failing with conflict - Fix
PostgreSQL: added retry logic for errors during upgrade task - Add
AlloyDBOmnifielduserConfig.node_count, typeinteger: Number of nodes for the service - Add
KafkafielduserConfig.kafka_diskless, typeobject: Kafka Diskless configuration values - Change
MySQLfielduserConfig.mysql.innodb_log_buffer_size: maximum→42949672954294967296 - Add
OpenSearchfielduserConfig.jwt, typeobject: OpenSearch JWT Configuration - Change
OpenSearchfielduserConfig.opensearch.knn_memory_circuit_breaker_limit: minimum→30 - Add
PostgreSQLfielduserConfig.node_count, typeinteger: Number of nodes for the service - Add
PostgreSQLfielduserConfig.pg.io_combine_limit, typeinteger: EXPERIMENTAL: Controls the
largest I/O size in operations that combine I/O in 8kB units - Add
PostgreSQLfielduserConfig.pg.io_max_combine_limit, typeinteger: EXPERIMENTAL: Controls
the largest I/O size in operations that combine I/O in 8kB units, and silently limits the user-settable
parameter io_combine_limit - Add
PostgreSQLfielduserConfig.pg.io_max_concurrency, typeinteger: EXPERIMENTAL: Controls the
maximum number of I/O operations that one process can execute simultaneously - Add
PostgreSQLfielduserConfig.pg.io_method, typestring: EXPERIMENTAL: Controls the maximum
number of I/O operations that one process can execute simultaneously - Add
PostgreSQLfielduserConfig.pg.io_workers, typeinteger: EXPERIMENTAL: Number of IO worker
processes, for io_method=worker. Version 18 and up only - Change
PostgreSQLfielduserConfig.pg.max_connections: maximum60000 - Remove
OpenSearchfielduserConfig.custom_keystores, typearray: the field has invalid schema definition - Add
KafkaTopicfieldconfig.diskless_enable, typeboolean: Indicates whether diskless should
be enabled - Remove
KafkaTopicfieldconfig.inkless_enable, typeboolean: Indicates whether inkless should
be enabled
v0.33.1
- Fix
PostgreSQL: resolved panic during upgrade check task fails
v0.31.1
- Fix
PostgreSQL: resolved panic during upgrade check task fails
v0.33.0
- DEPRECATION:
AlloyDBOmniis deprecated and entering its end-of-life cycle. See https://aiven.io/docs/platform/reference/end-of-life for details - Add
KafkafielduserConfig.kafka.log_message_timestamp_after_max_ms, typeinteger: The maximum
difference allowed between the timestamp when a broker receives a message and the timestamp specified
in the message - Add
KafkafielduserConfig.kafka.log_message_timestamp_before_max_ms, typeinteger: The maximum
difference allowed between the timestamp when a broker receives a message and the timestamp specified
in the message - Add
OpenSearchfielduserConfig.custom_keystores, typearray: Allow to register custom keystores
in OpenSearch - Change
OpenSearchfielduserConfig.opensearch_version: enum add2.19
v0.32.0
- BREAKING CHANGE: Removed unprefixed keys from ServiceUser secrets to resolve environment variable collisions. Previously ServiceUser secrets contained both prefixed keys (e.g.,
SERVICEUSER_HOST,SERVICEUSER_PASSWORD) and unprefixed keys (e.g.,HOST,PASSWORD). The unprefixed keys have been removed. - Important: Status conditions
Create,UpdateandCreateOrUpdate(all cases) have been consolidated intoCreatedOrUpdateddue to limitations in reliably determining operation type - Added secret watcher controller to automatically update resources when their
connInfoSecretSourcesecrets change - Fix
ServiceUser: retry API calls when password is not received in response - Add
AlloyDBOmnifielduserConfig.pg.max_sync_workers_per_subscription, typeinteger: Maximum
number of synchronization workers per subscription. The default is2 - Change
AlloyDBOmnifielduserConfig.pg.max_logical_replication_workers: maximum→64256 - Change
AlloyDBOmnifielduserConfig.pg.max_replication_slots: maximum→64256 - Change
AlloyDBOmnifielduserConfig.pg.max_worker_processes: maximum→96288 - Add
MySQLfielduserConfig.mysql_incremental_backup, typeobject: MySQL incremental backup configuration - Add
PostgreSQLfielduserConfig.pg.max_connections, typeinteger: Sets the PostgreSQL maximum
number of concurrent connections to the database server - Add
PostgreSQLfielduserConfig.pg.max_sync_workers_per_subscription, typeinteger: Maximum number
of synchronization workers per subscription. The default is2 - Change
PostgreSQLfielduserConfig.pg.max_logical_replication_workers: maximum→64256 - Change
PostgreSQLfielduserConfig.pg.max_replication_slots: maximum→64256 - Change
PostgreSQLfielduserConfig.pg.max_worker_processes: maximum→96288 - Change
ServiceIntegrationEndpointfieldprometheus.basic_auth_password: maxLength→64256 - Change
MySQLfielduserConfig.binlog_retention_period: maximum→86400604800
v0.31.0
KafkaTopic: added concurrent reconcilers and optimized API calls- Upgraded HPA from deprecated
autoscaling/v2beta1to stableautoscaling/v2API - Added
ServiceUserfieldconnInfoSecretSource: Allows reading passwords from existing secrets for credential management. Supports setting passwords for new users and existing users - Change
AlloyDBOmnifielduserConfig.pg.max_wal_senders: maximum→64256 - Add
KafkafielduserConfig.single_zone.availability_zone, typestring: The availability zone
to use for the service. This is only used when enabled is set to true - Change
PostgreSQLfielduserConfig.pg.max_wal_senders: maximum→64256 - Add
ClickhouseUserfieldconnInfoSecretSource: Allows reading passwords from existing secrets for credential management. Supports setting passwords for new users and existing users
v0.30.0
- Added
poweredfield (default:true) to control service power state. Whenfalse, the service is powered off.
Note: Kafka services without backups will lose topic data on power off. See field description for more information. - Completely replace the old go client with the new one, which is generated from the OpenAPI spec
- Change
PostgreSQLfielduserConfig.pg_version: enum remove12 - Add
KafkaTopicfieldconfig.inkless_enable, typeboolean: Indicates whether inkless should be enabled - Add
KafkaTopicfieldconfig.unclean_leader_election_enable, typeboolean: Indicates whether to
enable replicas not in the ISR set to be elected as leader as a last resort, even though doing so
may result in data loss - Refactor
KafkaTopic: replace HTTP client with code-generated one to improve maintainability and type safety - Add kind:
KafkaNativeACL. Creates and manages Kafka-native access control lists (ACLs) for an Aiven for Apache Kafka® service. - Add key
OPENSEARCH_URItoOpenSearchservice secrets: Contains the OpenSearch service URI. - Change
KafkaSchemafieldsschemaTypeandsubjectNameto be immutable since these fields cannot be modified after creation in the Kafka Schema Registry API - Improve
KafkaSchemacontroller: optimize polling and add better error handling - Improve
KafkaTopic: better handle API 5xx errors. - Improve
KafkaConnector: better handle API 404 and 5xx errors. - Fix webhooks
containerPortconfiguration not being properly applied in deployment template - Change
AlloyDBOmni,Cassandra,Clickhouse,Flink,Grafana,KafkaConnect,Kafka,MySQL,OpenSearch,PostgreSQL,Redis,ValkeyfielduserConfig.ip_filter: maxItems→20488000 - Add
ClickhousefielduserConfig.enable_ipv6, typeboolean: Register AAAA DNS records for the
service, and allow IPv6 packets to service ports - Add
OpenSearchfielduserConfig.opensearch.cluster.filecache.remote_data_ratio, typenumber:
Defines a limit of how much total remote data can be referenced as a ratio of the size of the disk
reserved for the file cache - Add
OpenSearchfielduserConfig.opensearch.cluster.remote_store, typeobject: no description - Add
OpenSearchfielduserConfig.opensearch.enable_snapshot_api, typeboolean: Enable/Disable
snapshot API for custom repositories, this requires security management to be enabled - Add
OpenSearchfielduserConfig.opensearch.node.search.cache.size, typestring: Defines a limit
of how much total remote data can be referenced as a ratio of the size of the disk reserved for
the file cache - Add
OpenSearchfielduserConfig.opensearch.remote_store, typeobject: no description
v0.29.0
- Added retry logic to the
ServiceIntegrationcontroller - Made
ConnectionPoolusername field optional, allowing connection pools to use the credentials of the connecting client instead of a fixed service user - Add
KafkafielduserConfig.kafka_rest_config.consumer_idle_disconnect_timeout, typeinteger:
Specifies the maximum duration (in seconds) a client can remain idle before it is deleted - Change
ServiceIntegrationfieldclickhouseKafka.tables: maxItems→100400 - Add
ValkeyfielduserConfig.enable_ipv6, typeboolean: Register AAAA DNS records for the service,
and allow IPv6 packets to service ports - Add
ValkeyfielduserConfig.valkey_active_expire_effort, typeinteger: Valkey reclaims expired
keys both when accessed and in the background - Add
OpenSearchfielduserConfig.azure_migration.readonly, typeboolean: Whether the repository
is read-only - Add
OpenSearchfielduserConfig.gcs_migration.readonly, typeboolean: Whether the repository
is read-only - Add
OpenSearchfielduserConfig.opensearch.disk_watermarks, typeobject: Watermark settings - Add
OpenSearchfielduserConfig.s3_migration.readonly, typeboolean: Whether the repository is
read-only - Add
AlloyDBOmnifielduserConfig.pgaudit, typeobject: System-wide settings for the pgaudit extension - Add
ClickhousefielduserConfig.backup_hour, typeinteger: The hour of day (in UTC) when backup
for the service is started - Add
ClickhousefielduserConfig.backup_minute, typeinteger: The minute of an hour when backup
for the service is started - Add
KafkafielduserConfig.kafka_connect_plugin_versions, typearray: The plugin selected by the user - Change
KafkafielduserConfig.kafka_version: enum add3.9 - Add
OpenSearchfielduserConfig.opensearch.enable_searchable_snapshots, typeboolean: Enable
searchable snapshots - Change
PostgreSQLfielduserConfig.pgaudit.log_level: enum adddebug1,debug2,debug3,debug4,
debug5,info,log,notice
v0.28.0
- Add kind:
AlloyDBOmni - Deprecate
Redis: useValkeyinstead. Please follow these instructions to upgrade your service to Valkey - Deprecate
Cassandra, see Aiven platform end-of-life policy. - Change
CassandrafielduserConfig.ip_filter: maxItems→10242048 - Change
ClickhousefielduserConfig.ip_filter: maxItems→10242048 - Change
FlinkfielduserConfig.custom_code: immutabletrue - Change
FlinkfielduserConfig.ip_filter: maxItems→10242048 - Add
GrafanafielduserConfig.dashboard_scenes_enabled, typeboolean: Enable use of the Grafana
Scenes Library as the dashboard engine. i.e - Change
GrafanafielduserConfig.ip_filter: maxItems→10242048 - Add
KafkaConnectfielduserConfig.plugin_versions, typearray: The plugin selected by the user - Change
KafkaConnectfielduserConfig.ip_filter: maxItems→10242048 - Change
KafkafielduserConfig.ip_filter: maxItems→10242048 - Change
MySQLfielduserConfig.ip_filter: maxItems→10242048 - Add
OpenSearchfielduserConfig.opensearch.cluster.search.request.slowlog, typeobject - Add
OpenSearchfielduserConfig.opensearch.enable_remote_backed_storage, typeboolean: Enable
remote-backed storage - Change
OpenSearchfielduserConfig.ip_filter: maxItems→10242048 - Change
PostgreSQLfielduserConfig.ip_filter: maxItems→10242048 - Change
RedisfielduserConfig.ip_filter: maxItems→10242048 - Change
ServiceIntegrationfieldlogs.elasticsearch_index_prefix: pattern^[a-z0-9][a-z0-9-_.]+$ - Change
ValkeyfielduserConfig.ip_filter: maxItems→10242048 - Add
ValkeyfielduserConfig.frequent_snapshots, typeboolean: When enabled, Valkey will create
frequent local RDB snapshots - Change
OpenSearchfielduserConfig.opensearch.auth_failure_listeners.internal_authentication_backend_limiting.allowed_tries:
maximum→214748364732767 - Change
OpenSearchfielduserConfig.opensearch.auth_failure_listeners.ip_rate_limiting: deprecated - Add
DatabasefielddatabaseNametypestring: DatabaseName is the name of the database to be created.
v0.27.0
- Add
ServiceIntegrationEndpointfielddatadog.extra_tags_prefix, typestring: Extra tags prefix.
Defaults to aiven - Change
FlinkfielduserConfig.flink_version: enum add1.20 - Add
OpenSearchfielduserConfig.opensearch_dashboards.multiple_data_source_enabled, typeboolean:
Enable or disable multiple data sources in OpenSearch Dashboards - Change
OpenSearchfielduserConfig.opensearch_dashboards.max_old_space_size: maximum2048
→4096 - Change
PostgreSQLfielduserConfig.pg_version: enum add17 - Add
PostgreSQLfielduserConfig.pg.password_encryption, typestring: Chooses the algorithm for
encrypting passwords - Add
OpenSearchfielduserConfig.opensearch.cluster.routing.allocation.balance.prefer_primary, type
boolean: When set to true, OpenSearch attempts to evenly distribute the primary shards between
the cluster nodes - Add
OpenSearchfielduserConfig.opensearch.segrep, typeobject: Segment Replication Backpressure
Settings - Add
FlinkfielduserConfig.custom_code, typeboolean: Enable to upload Custom JARs for Flink
applications - Add kind:
Valkey