It has been such a wonderful week so far because another tool under the CNCF umbrella got its latest update. There are many changes, bug fixes, and new features and enhancements in this new update. We are going to talk about all of these in this article. But, let’s get around with Keda at first!
What is KEDA?
KEDA is a Cloud Native Computing Foundation (CNCF) sandbox project that allows fine-grained autoscaling (including to/from zero) for event-driven Kubernetes workloads. KEDA is known for serving as a Kubernetes Metrics Server, and it will enable users to define autoscaling rules using a dedicated Kubernetes custom resource definition. KEDA can run both on the cloud and the edge. It also integrates natively with Kubernetes components such as the Horizontal Pod Autoscaler and has no external dependencies.
So, we are going to see what the new update brought. Let’s start.
Highlights of 2.4.0
The new version has a lot of highlighting features that need mentioning. With version 2.4.0, we see the introduction of a new Solace PubSub+ Event Broker scaler. Again, we can see the introduction of a new Selenium Grid scaler and a new Kubernetes Workload scaler. This version also comes with a significant release of fallback functionality and Idle Replica Mode. You can learn how to deploy Keda by reading this.
All of the features that we have seen in the highlighting section are new features. Apart from those, we can see another essential feature with this new update: ScaledJob. This new feature will support pod conditions for pending job count calculation from now on.
With the new version, we can see the optimization of the Kafka scaler by fetching all topic offsets using a single HTTP request. Also, we see the adding ability to specify Kafka Broker Version. Version 2.4.0 comes with the support of a custom metric name in RabbitMQ scaler. It also now supports the use of regex to select the queues in RabbitMQ scaler.
Another significant improvement came with the extension of the Azure Monitor scaler to support custom metrics. Also, speaking of Azure, some more modifications like supporting non-public cloud environments in the Azure Service Bus scaler is available. Another Azure improvement like supporting non-public cloud environments in the Azure Storage Queue and Azure Storage Blob scalers will now become visible.
The new update comes with an adjustment of the InfluxDB scaler, which will support queries that return integers and in addition to those that return floats. It will now allow sourcing of InfluxDB
(Cluster)TriggerAuthentication. An essential improvement like the IBM MQ scaler password handling fix will make Keda perform better.
Another significant improvement with Metrics APIServer, which will now add rate-limiting parameters to override the client. The new update also comes with fixing READY and ACTIVE fields of ScaledJob to show status when we run
kubectl get sj. It also indicates HashiCorp Vault Address when using
kubectl get ta or
kubectl get cta. Specific improvements come with the new version where we don’t have to panic when the HashiCorp Vault path doesn’t exist.
The breaking change which came along with this new update is the fixing of the
keda-system-auth-delegator ClusterRoleBinding name. The Upgradation may leave a stray ClusterRoleBinding with the old name
The only other change we see in version 2.4.0 is using
scaled[object/job].keda.sh/ prefix for KEDA related labels.
Throughout the article, we have seen the improvements and new features that come with this new update. It all makes KEDA easier to handle, along with more user engagement. You can also try out this awesomeness and get the latest update by clicking here.
Join our inclusive and developer centeric event here.
Feel free to read our other posts: