Skip to content

Latest commit

 

History

History
94 lines (69 loc) · 3.89 KB

durable-function.md

File metadata and controls

94 lines (69 loc) · 3.89 KB
Raw

Durable Functions

Durable Functions is an extension of Azure Functions that enables you to perform long-lasting, stateful operations in Azure. Azure provides the infrastructure for maintaining state information. You can use Durable Functions to orchestrate a long-running workflow. Using this approach, you get all the benefits of a serverless hosting model, while letting the Durable Functions framework take care of activity monitoring, synchronization, and runtime concerns.

Tutorial

1. Create a new directory for the function app

mkdir DurableFunction
cd DurableFunction

2. Initialize the directory for functions

func init . --docker

Select dotnet

3. Add a new durable function

func new

Select DurableFunctionsOrchestration then select a name for your function.

4. Run function

func start

What is KEDA Durable Functions Scaler?

KEDA supports multiple scalers. As a part of the scalers, this project support Durable Functions Scaler for KEDA. You can deploy Durable Functions with auto scale feature on Kubernetes.

How KEDA Durable Functions Scaler works

KEDA Durable Functions Scaler works as a gRPC server of the External Scaler Support.

Overview

Limitations

Minimum Pod number is 1. Not zero.

Currently, KEDA Durable Scaler can't make functions scale down to zero. The minimum pod number is one. Durable Scaler needs to send data to the control/worker queue. To achieve this behavior, we need to separate the HTTP and non-HTTP deployments. However, the feature seems not working. We need to wait until this issue is fixed.

Keda yaml file

for deploy keda on your kubernetes cluster the yaml file would be like this (see complete file):

apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
  name: azure-durable-function-with-keda
  namespace: keda-namespace
spec:
  scaleTargetRef:
    name: azure-durable-function-with-keda
  minReplicaCount: 1
  maxReplicaCount: 20
  triggers:
    - type: external
      metadata:
        scalerAddress: durable-function-address #Required and can be public address or internal address

there are some important things in KEDA to be explain

  • kind: ScaledObject tells kubernetes that this service is for keda and use for auto scalling
  • scaleTargetRef consider that KEDA should trigger which service that run on kubernetes for detecting auto scalling when service was busy and needs more pods (in this case we first create deployment named durable-function)
  • triggers consider type of scaled object, there several trigger types, for Durable function use external
  • scalerAddress this is the address that KEDA will check it for increasing or decreasing pods

Having all that configured we can now deploy the YAML file to AKS.

kubectl apply -f .\durable-function-deploy.yml

To check if everything is working, check the deployment.

> kubectl get deploy
NAME                        READY   UP-TO-DATE   AVAILABLE   AGE
 azure-durable-function-with-keda            0/0     0            0           20h

> kubectl get ScaledObject
NAME                       SCALETARGETKIND           SCALETARGETNAME             TRIGGERS      AUTHENTICATION   READY   ACTIVE      AGE
azure-durable-function-with-keda   apps/v1.Deployment   azure-durable-function-with-keda            external                       True      False     20h