Skip to content

Releases: kedacore/keda

v2.16.1

24 Dec 19:28
ce14b23
Compare
Choose a tag to compare

We are happy to release KEDA v2.16.1 🎉

Here are some fixes:

  • General: Centralize and improve automaxprocs configuration with proper structured logging (#5970)
  • General: Paused ScaledObject count is reported correctly after operator restart (#6321)
  • General: ScaledJobs ready status set to true when recoverred problem (#6329)
  • Selenium Grid Scaler: Exposes sum of pending and ongoing sessions to KDEA (#6368)
  • Selenium Grid Scaler: Selenium Grid in case multiple scaler triggers are activate (#6437)

Learn how to deploy KEDA by reading our documentation.

Full Changelog: v2.16.0...v2.16.1

v2.16.0

07 Nov 09:47
5c52d03
Compare
Choose a tag to compare

We are happy to release KEDA v2.16.0 🎉

Here are some highlights:

  • Introduce new Beanstalkd scaler (#5901)
  • Introduce ClusterCloudEventSource (#3533)
  • Replace wildcards in RBAC objects with explicit resources and verbs (#6129)
  • Multiple scaler improvements

Here are the new breaking change(s) of this release:

  • Pulsar Scaler: remove msgBacklog trigger name (#6059)

Here are the new deprecation(s) as of this release:

  • IBM MQ Scaler: Remove and deprecate unused variables in IBM MQ scaler (#6033)

Learn how to deploy KEDA by reading our documentation.

🗓️ The next KEDA release is currently being estimated for Jan 30th, 2025, learn more in our roadmap.

New Contributors

Full Changelog: v2.15.0...v2.16.0

v2.15.1

12 Aug 10:30
09a4951
Compare
Choose a tag to compare

We are happy to release KEDA v2.15.1 🎉

Here are some fixes:

  • General: Hashicorp Vault PKI doesn't fail with due to KeyPair mismatch (#6028)
  • JetStream: Consumer leader change is correctly detected (#6042)

Learn how to deploy KEDA by reading our documentation.

v2.15.0

01 Aug 12:07
3aa0f7e
Compare
Choose a tag to compare

We are happy to release KEDA v2.15.0 🎉

Here are some highlights:

  • Add Dynatrace Scaler (#5685)
  • Add Splunk Scaler (#5904)
  • Fix panic in KedaProvider when getting metrics from Metrics Service if the gRPC Server connection is not established
  • Scalers are properly closed after being refreshed

Here are the new breaking change(s) as of this release:

  • AAD-Pod-Identity and AWS-KIAM auths have been removed (#5035|#5085)
  • Remove deprecated field cortexOrgID from Prometheus scaler (#5538)

Learn how to deploy KEDA by reading our documentation.

New Contributors

Full Changelog: v2.14.0...v2.15.0

v2.14.1

31 Jul 15:17
10840ef
Compare
Choose a tag to compare

We are happy to release KEDA v2.14.1 🎉

Here are some fixes:

  • General: Do not delete running Jobs on KEDA restart (#5656)
  • General: Fix CVE-2024-24790, CVE-2024-24789, and CVE-2024-24791 in stdlib. (#5971)
  • General: Fix CVE-2024-35255 in github.com/Azure/azure-sdk-for-go/sdk/azidentity (#5971)
  • General: Fix ScaledJob ignores failing trigger(s) error (#5922)- General: Scalers are properly closed after being refreshed (#5806)
  • GitHub Scaler: Fixed pagination, fetching repository list (#5738)
  • MongoDB Scaler: MongoDB url parses correctly +srv scheme (#5760)

v2.14.0

25 Apr 13:02
6681d5e
Compare
Choose a tag to compare

We are happy to release KEDA v2.14.0 🎉

Here are some highlights:

  • Provide capability to filter CloudEvents (#3533)
  • (Experimental) Introduce Azure Event Grid as a new CloudEvent destination (#3587)
  • Support Kafka SASL MSK IAM authentication
  • Multiple OpenTelemetry improvements
  • Multiple ScalingModifiers improvements and fixes

Here are the new deprecation(s) as of this release:

  • Various Prometheus metrics have been renamed to follow the preferred naming conventions. The old ones are still available, but will be removed in the future (#4854).

Learn how to deploy KEDA by reading our documentation.

🗓️ The next KEDA release is currently being estimated for July 9th, 2024, learn more in our roadmap.

New Contributors

Full Changelog: v2.13.0...v2.14.0

v2.13.1

29 Feb 16:33
41dd7a8
Compare
Choose a tag to compare

We are happy to release KEDA v2.13.1 🎉

Here are some fixes:

  • General: Fix release asset should specify the version in keda-*-core.yaml(#5484)
  • GCP Scalers: Properly close the connection during the scaler cleaning process (#5448)
  • GCP Scalers: Restore previous time horizon to prevent querying issues (#5429)
  • Prometheus Scaler: Fix for missing AWS region from metadata (#5419)

What's Changed

Full Changelog: v2.13.0...v2.13.1

v2.13.0

19 Jan 06:05
367fcd3
Compare
Choose a tag to compare

We are happy to release KEDA v2.13 🎉

Here are some highlights:

  • Support for GCP Secret Manager & config map in trigger authentication
  • Newly improved AWS authentication
  • Support for emitting CloudEvents from KEDA allowing end-users to extend KEDA (Experimental)
  • Support for Amazon managed service for Prometheus for Prometheus scaler
  • SAS token authentication for Azure Storage scalers
  • Workload identity authentication for Azure Pipelines
  • Additional metrics and performance improvements

Here are the breaking changes of this release:

  • General: Clean up previously deprecated code in Azure Data Explorer Scaler about clientSecret for 2.13 release (#5051)

Here are the new deprecation(s) as of this release:

  • Remove support for Azure AD Pod Identity-based authentication (#5035)

Learn how to deploy KEDA by reading our documentation.

🗓️ The next KEDA release is currently being estimated for April 12th, 2024, learn more in our roadmap.

New

  • General: Adds support for GCP Secret Manager as a source for TriggerAuthentication (#4831)
  • General: Support TriggerAuthentication properties from ConfigMap (#4830)
  • General: Introduce new AWS Authentication (#4134)
  • Azure Blob Storage Scaler: Allow to authenticate to Azure Storage using SAS tokens (#5393)
  • Azure Pipelines Scaler: Add support for workload identity authentication (#5013)
  • Azure Storage Queue Scaler: Allow to authenticate to Azure Storage using SAS tokens (#5393)
  • Kafka Scaler: Add support for Kerberos authentication (SASL / GSSAPI) (#4836)
  • Prometheus Metrics: Expose prometheus metrics for ScaledJob resources (#4798)
  • Prometheus Metrics: Introduce paused ScaledObjects in Prometheus metrics (#4430)
  • Prometheus Scaler: Provide scaler for Amazon managed service for Prometheus (#2214)

Experimental

Here is an overview of all new experimental features:

  • General: Emit CloudEvents on major KEDA events (#3533|#5278)

Improvements

  • General: Add CloudEventSource metrics in Prometheus & OpenTelemetry (#3531)
  • General: Add RBAC permissions for list & watch LimitRange, and check default limits from LimitRange for validations (#5377)
  • General: Add validations for replica counts when creating ScaledObjects (#5288)
  • General: Bubble up AuthRef TriggerAuthentication errors as ScaledObject events (#5190)
  • General: Enhance pod identity role assumption in AWS by directly integrating with OIDC/Federation (#5178)
  • General: Fix issue where paused annotation being set to false still leads to ScaledObjects/ScaledJobs being paused (#5215)
  • General: Implement credentials cache for AWS Roles to reduce AWS API calls (#5297)
  • General: Request all ScaledObject/ScaledJob triggers in parallel (#5276)
  • General: Use client-side round-robin load balancing for gRPC calls (#5224)
  • GCP PubSub Scaler: Support distribution-valued metrics and metrics from topics (#5070)
  • GCP Stackdriver Scaler: Support valueIfNull parameter (#5345)
  • Hashicorp Vault: Add support to get secret that needs write operation (eg. pki) (#5067)
  • Hashicorp Vault: Fix operator panic when spec.hashiCorpVault.credential.serviceAccount is not set (#4964)
  • Hashicorp Vault: Fix operator panic when using root token to authenticate to vault server (#5192)
  • Kafka Scaler: Ability to set upper bound to the number of partitions with lag (#3997)
  • Kafka Scaler: Improve logging for Sarama client (#5102)
  • Prometheus Scaler: Add queryParameters parameter (#4962)
  • Pulsar Scaler: Support `endpointParams`` in Pulsar OAuth (#5069)

Fixes

  • General: Admission webhook does not reject workloads with only resource limits provided (#4802)
  • General: Fix CVE-2023-39325 in golang.org/x/net (#5122)
  • General: Fix otelgrpc DoS vulnerability (#5208)
  • General: Fix Pod identity not being considered when scaled target is a CRD (#5021)
  • General: Prevented memory leak generated by not correctly cleaning http connections (#5248)
  • General: Prevented stuck status due to timeouts during scalers generation (#5083)
  • General: ScaledObject Validating Webhook should support dry-run=server requests (#5306)
  • General: Set LeaderElectionNamespace to PodNamespace so leader election works in OutOfCluster mode (#5404)
  • AWS Scalers: Ensure session tokens are included when instantiating AWS credentials (#5156)
  • Azure Event Hub Scaler: Improve unprocessedEventThreshold calculation (#4250)
  • Azure Pipelines: Prevent HTTP 400 errors due to poolName with spaces (#5107)
  • GCP PubSub Scaler: Added project_id to filter for metrics queries (#5256)
  • GCP PubSub Scaler: Respect default value of value (#5093)
  • Github Runner Scaler: Support for custom API endpoint (#5387)
  • NATS JetSteam Scaler: Raise an error if leader not found (#5358)
  • Pulsar Scaler: Fix panic when auth is not used (#5271)
  • ScaledJobs: Copy ScaledJob annotations to child Jobs (#4594)

Deprecations

You can find all deprecations in this overview and join the discussion here.

New deprecation(s):

  • Remove support for Azure AD Pod Identity-based authentication (#5035)

Breaking Changes

  • General: Clean up previously deprecated code in Azure Data Explorer Scaler about clientSecret for 2.13 release (#5051)

Other

  • General: Bump K8s deps to 0.28.5 (#5346)
  • General: Create a common utility function to get parameter value from config (#5037)
  • General: Fix CVE-2023-45142 in OpenTelemetry (#5089)
  • General: Fix logger in OpenTelemetry collector (#5094)
  • General: Fix lost commit from the newly created utility function (#5037)
  • General: Improve docker image build time through caches (#5316)
  • General: Reduce amount of gauge creations for OpenTelemetry metrics (#5101)
  • General: Refactor scalers package (#5379)
  • General: Removed not required RBAC permissions (#5261)
  • General: Support profiling for KEDA components (#4789)
  • CPU scaler: Wait for metrics window during CPU scaler tests (#5294)
  • Hashicorp Vault: Improve test coverage in pkg/scaling/resolver/hashicorpvault_handler (#5195)
  • Kafka Scaler: Add more test cases for large value of LagThreshold (#5354)
  • Openstack Scaler: Use Gophercloud SDK (#3439)

New Contributors

Read more

KEDA v2.12.1

27 Nov 21:18
Compare
Choose a tag to compare

KEDA 2.12.1

We are happy to release 2.12.1 🎉

Learn how to deploy KEDA by reading our documentation.

🗓️ The next KEDA release is currently being estimated for January 11th, 2024 (planned later due to holidays), learn more in our roadmap.

Fixes

  • General: Fix CVE-2023-39325 in golang.org/x/net (#5122)
  • General: Fix CVE-2023-45142 in Opentelemetry (#5089)
  • General: Fix logger in Opentelemetry collector (#5094)
  • General: Fix otelgrpc DoS vulnerability (#5208)
  • General: Prevented stuck status due to timeouts during scalers generation (#5083)
  • Azure Pipelines: No more HTTP 400 errors produced by poolName with spaces (#5107)

What's Changed

Full Changelog: v2.12.0...v2.12.1

KEDA v2.12.0

28 Sep 15:39
9527a7f
Compare
Choose a tag to compare

We are happy to release KEDA v2.12 🎉

Here are some highlights:

  • New Google Cloud Tasks scaler
  • New apache-kafka scaler (Experimental)
  • Support for formula-based evaluation of metric values (Experimental)
  • Support for pushing operational metrics to an OpenTelemetry Collector (Experimental)

This release has no new deprecations.

Learn how to deploy KEDA by reading our documentation.

🗓️ The next KEDA release is currently being estimated for January 11th, 2024 (planned later due to holidays), learn more in our roadmap.

New

  • General: Introduce new Google Cloud Tasks scaler (#3613)
  • AWS SQS Scaler: Support for scaling to include delayed messages. (#4377)
  • Governance: KEDA transitioned to CNCF Graduated project (#63)

Experimental

Here is an overview of all new experimental features:

  • General: New apache-kafka scaler based on kafka-go library (#4692)
  • General: Support for formula-based evaluation of metric values (#2440|#4998)
  • General: Introduce pushing operational metrics to an OpenTelemetry Collector (#3078)

Improvements

  • General: Introduce Prometheus metrics of API server to KEDA Metric Server (#4460)
  • General: Show ScaledObject/ScaledJob names to output of kubectl get triggerauthentication/clustertriggerauthentication (#796)
  • General: Introduce standalone CRD generation to release workflow (#2726)
  • General: Introduce changelog validation check during PR for formatting and order violations (#3190)
  • General: Automatically set GOMAXPROCS to match Linux container CPU quota (#4999)
  • General: Configurable Kubernetes cluster domain for Cert Manager (#4861)
  • General: Provide new Kubernetes events about internal state and scaling (#3764)
  • General: Introduce annotation autoscaling.keda.sh/paused: true for ScaledObject to pause autoscaling (#3304)
  • General: Updated AWS SDK and updated all the AWS scalers (#4905)
  • Azure Pod Identity: Introduce validation to prevent usage of empty identity ID for Azure identity providers (#4528)
  • Kafka Scaler: Add unsafeSsl flag for kafka scaler (#4977)
  • Prometheus Scaler: Remove trailing whitespaces in customAuthHeader and customAuthValue (#4960)
  • Pulsar Scaler: Support for OAuth extensions (#4700)
  • Redis Scalers: Support for TLS authentication for Redis & Redis stream scalers (#4917)

Fixes

  • General: Add validations for stabilizationWindowSeconds (#4976)
  • RabbitMQ Scaler: Allow subpaths along with vhost in connection string (#2634)
  • Selenium Grid Scaler: Fix scaling based on latest browser version (#4858)
  • Solace Scaler: Fix a bug where queueName is not properly escaped during URL encode (#4936)

You can find all deprecations in this overview and join the discussion here.

New deprecation(s):

  • None.

Breaking Changes

  • General: Remove deprecated metricName from trigger metadata section (#4899)

Other

  • General: Fixed a typo in the StatefulSet scaling resolver (#4902)
  • General: Only show logs with a severity level of ERROR or higher in the stderr in metrics server (#4049)
  • General: Refactor ScaledJob related methods to be located at scale_handler (#4781)
  • General: Replace deprecated set-output command with environment file in GitHub Actions workflows (#4914)

New Contributors

Full Changelog: v2.11.0...v2.12.0