Manage orchestration instances

[FunctionName("HelloWorldQueueTrigger")]
public static async Task Run(
    [QueueTrigger("start-queue")] string input,
    [DurableClient] IDurableOrchestrationClient starter,
    ILogger log)
{
    string instanceId = await starter.StartNewAsync("HelloWorld", input);
    log.LogInformation($"Started orchestration with ID = '{instanceId}'.");
}

Unless otherwise specified, the examples on this page use the HTTP trigger with the following function.json.

function.json

{
  "bindings": [
    {
      "name": "req",
      "type": "httpTrigger",
      "direction": "in",
      "methods": ["post"]
    },
    {
      "name": "$return",
      "type": "http",
      "direction": "out"
    },
    {
      "name": "starter",
      "type": "durableClient",
      "direction": "in"
    }
  ],
  "disabled": false
}

index.js

const df = require("durable-functions");

module.exports = async function(context, input) {
    const client = df.getClient(context);

    const instanceId = await client.startNew("HelloWorld", undefined, input);
    context.log(`Started orchestration with ID = ${instanceId}.`);
};

Unless otherwise specified, the examples on this page use the HTTP trigger with the following function.json.

function.json

{
  "scriptFile": "__init__.py",
  "bindings": [    
    {
      "name": "req",
      "type": "httpTrigger",
      "direction": "in",
      "methods": ["post"]
    },
    {
      "name": "$return",
      "type": "http",
      "direction": "out"
    },
    {
      "name": "starter",
      "type": "durableClient",
      "direction": "in"
    }
  ],
  "disabled": false
}

__init__.py

import logging
import azure.functions as func
import azure.durable_functions as df

async def main(req: func.HttpRequest, starter: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)

    instance_id = await client.start_new('HelloWorld', None, None)
    logging.log(f"Started orchestration with ID = ${instance_id}.")

Unless otherwise specified, the examples on this page use the HTTP trigger with the following function.json.

function.json

{
  "bindings": [    
    {
      "name": "Request",
      "type": "httpTrigger",
      "direction": "in",
      "methods": ["post"]
    },
    {
      "name": "Response",
      "type": "http",
      "direction": "out"
    },
    {
      "name": "starter",
      "type": "durableClient",
      "direction": "in"
    }
  ],
  "disabled": false
}

run.ps1

param($Request, $TriggerMetadata)

$InstanceId = Start-DurableOrchestration -FunctionName 'HelloWorld'
Write-Host "Started orchestration with ID = '$InstanceId'"

@FunctionName("HelloWorldQueueTrigger")
public void helloWorldQueueTrigger(
        @QueueTrigger(name = "input", queueName = "start-queue", connection = "Storage") String input,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext,
        final ExecutionContext context) {
    DurableTaskClient client = durableContext.getClient();        
    String instanceID = client.scheduleNewOrchestrationInstance("HelloWorld");
    context.getLogger().info("Scheduled orchestration with ID = " + instanceID);
}

To wait for the orchestrator to start before returning from your function, use the waitForInstanceStart() method.

// wait up to 30 seconds for the scheduled orchestration to enter the "Running" state
client.waitForInstanceStart(instanceID, Duration.ofSeconds(30));

using Microsoft.DurableTask.Client;

// Schedule a new orchestration instance
string instanceId = await client.ScheduleNewOrchestrationInstanceAsync("HelloWorld", input);
Console.WriteLine($"Started orchestration with ID = '{instanceId}'.");

// Optionally, wait for the orchestration to start
OrchestrationMetadata metadata = await client.WaitForInstanceStartAsync(instanceId, timeout: TimeSpan.FromSeconds(30));

from durabletask.azuremanaged.client import DurableTaskSchedulerClient

# Schedule a new orchestration instance
instance_id = client.schedule_new_orchestration(hello_world, input=input_data)
print(f"Started orchestration with ID = '{instance_id}'.")

# Optionally, wait for the orchestration to start
state = client.wait_for_orchestration_start(instance_id, timeout=30)

import com.microsoft.durabletask.DurableTaskClient;

// Schedule a new orchestration instance
String instanceId = client.scheduleNewOrchestrationInstance("HelloWorld", input);
System.out.println("Started orchestration with ID = '" + instanceId + "'.");

// Optionally, wait for the orchestration to start
OrchestrationMetadata metadata = client.waitForInstanceStart(instanceId, Duration.ofSeconds(30), false);

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

// Schedule a new orchestration instance
const instanceId = await client.scheduleNewOrchestration("HelloWorld", input);
console.log(`Started orchestration with ID = '${instanceId}'.`);

// Optionally, wait for the orchestration to start
const state = await client.waitForOrchestrationStart(instanceId, false, 30);

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

[FunctionName("GetStatus")]
public static async Task Run(
    [DurableClient] IDurableOrchestrationClient client,
    [QueueTrigger("check-status-queue")] string instanceId)
{
    DurableOrchestrationStatus status = await client.GetStatusAsync(instanceId);
    // do something based on the current status.
}

const df = require("durable-functions");

module.exports = async function(context, instanceId) {
    const client = df.getClient(context);

    const status = await client.getStatus(instanceId);
    // do something based on the current status.
    // example: if status.runtimeStatus === df.OrchestrationRuntimeStatus.Running: ...
}

For the function.json configuration, see Start instances.

import azure.functions as func
import azure.durable_functions as df

async def main(req: func.HttpRequest, starter: str, instance_id: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)

    status = await client.get_status(instance_id)
    # do something based on the current status
    # example: if (existing_instance.runtime_status is df.OrchestrationRuntimeStatus.Running) { ...

param($Request, $TriggerMetadata)

# Get instanceid from body
$InstanceId = $Request.Body.InstanceId

$Status = Get-DurableStatus -InstanceId $InstanceId -ShowHistory -ShowHistoryOutput -ShowInput
Write-Host "Status: $($Status | ConvertTo-Json)"

# Do something based on status
# example: if ($Status.runtimeStatus -eq 'Running') { ... }

@FunctionName("GetStatus")
public void getStatus(
        @QueueTrigger(name = "instanceID", queueName = "check-status-queue", connection = "Storage") String instanceID,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext,
        final ExecutionContext context) {
    DurableTaskClient client = durableContext.getClient();        
    OrchestrationMetadata metadata = client.getInstanceMetadata(instanceID, false);
    if (metadata != null) {
        OrchestrationRuntimeStatus status = metadata.getRuntimeStatus();
        switch (status) {
            // do something based on the current status
        }
    }
}

using Microsoft.DurableTask.Client;

// Get the status of an orchestration instance
OrchestrationMetadata? metadata = await client.GetInstanceAsync(instanceId, getInputsAndOutputs: true);
if (metadata != null)
{
    OrchestrationRuntimeStatus status = metadata.RuntimeStatus;
    // do something based on the current status
}

from durabletask.azuremanaged.client import DurableTaskSchedulerClient

# Get the status of an orchestration instance
state = client.get_orchestration_state(instance_id, fetch_payloads=True)
if state is not None:
    status = state.runtime_status
    # do something based on the current status

import com.microsoft.durabletask.DurableTaskClient;
import com.microsoft.durabletask.OrchestrationMetadata;

// Get the status of an orchestration instance
OrchestrationMetadata metadata = client.getInstanceMetadata(instanceId, true);
if (metadata != null) {
    OrchestrationRuntimeStatus status = metadata.getRuntimeStatus();
    // do something based on the current status
}

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

// Get the status of an orchestration instance
const state = await client.getOrchestrationState(instanceId, true);
if (state) {
    const status = state.runtimeStatus;
    // do something based on the current status
}

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

[FunctionName("GetAllStatus")]
public static async Task Run(
    [HttpTrigger(AuthorizationLevel.Anonymous, "get", "post")] HttpRequestMessage req,
    [DurableClient] IDurableOrchestrationClient client,
    ILogger log)
{
    var noFilter = new OrchestrationStatusQueryCondition();
    OrchestrationStatusQueryResult result = await client.ListInstancesAsync(
        noFilter,
        CancellationToken.None);
    foreach (DurableOrchestrationStatus instance in result.DurableOrchestrationState)
    {
        log.LogInformation(JsonConvert.SerializeObject(instance));
    }

    // Note: ListInstancesAsync only returns the first page of results.
    // To request additional pages provide the result.ContinuationToken
    // to the OrchestrationStatusQueryCondition's ContinuationToken property.
}

const df = require("durable-functions");

module.exports = async function(context, req) {
    const client = df.getClient(context);

    const instances = await client.getStatusAll();
    instances.forEach((instance) => {
        context.log(JSON.stringify(instance));
    });
};

For the function.json configuration, see Start instances.

import logging
import json
import azure.functions as func
import azure.durable_functions as df


async def main(req: func.HttpRequest, starter: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)

    instances = await client.get_status_all()

    for instance in instances:
        logging.log(json.dumps(instance))

For the function.json configuration, see Start instances.

@FunctionName("GetAllStatus")
public String getAllStatus(
        @HttpTrigger(name = "req", methods = {HttpMethod.GET}) HttpRequestMessage<?> req,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext) {
    DurableTaskClient client = durableContext.getClient();
    OrchestrationStatusQuery noFilter = new OrchestrationStatusQuery();
    OrchestrationStatusQueryResult result = client.queryInstances(noFilter);
    return "Found " + result.getOrchestrationState().size() + " orchestrations.";
}

using Microsoft.DurableTask.Client;

// Query all orchestration instances
AsyncPageable<OrchestrationMetadata> instances = client.GetAllInstancesAsync(new OrchestrationQuery());

await foreach (OrchestrationMetadata instance in instances)
{
    Console.WriteLine(instance.InstanceId);
}

from durabletask.azuremanaged.client import DurableTaskSchedulerClient

# Query all orchestration instances
instances = client.list_orchestrations()

for instance in instances:
    print(instance.instance_id)

import com.microsoft.durabletask.DurableTaskClient;
import com.microsoft.durabletask.OrchestrationStatusQuery;
import com.microsoft.durabletask.OrchestrationStatusQueryResult;

// Query all orchestration instances
OrchestrationStatusQuery query = new OrchestrationStatusQuery();
OrchestrationStatusQueryResult result = client.queryInstances(query);

for (OrchestrationMetadata instance : result.getOrchestrationState()) {
    System.out.println(instance.getInstanceId());
}

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

// Query all orchestration instances
const instances = client.getAllInstances();

for await (const instance of instances) {
    console.log(instance.instanceId);
}

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

[FunctionName("QueryStatus")]
public static async Task Run(
    [HttpTrigger(AuthorizationLevel.Anonymous, "get", "post")] HttpRequestMessage req,
    [DurableClient] IDurableOrchestrationClient client,
    ILogger log)
{
    // Get the first 100 running or pending instances that were created between 7 and 1 days ago
    var queryFilter = new OrchestrationStatusQueryCondition
    {
        RuntimeStatus = new[]
        {
            OrchestrationRuntimeStatus.Pending,
            OrchestrationRuntimeStatus.Running,
        },
        CreatedTimeFrom = DateTime.UtcNow.Subtract(TimeSpan.FromDays(7)),
        CreatedTimeTo = DateTime.UtcNow.Subtract(TimeSpan.FromDays(1)),
        PageSize = 100,
    };

    OrchestrationStatusQueryResult result = await client.ListInstancesAsync(
        queryFilter,
        CancellationToken.None);
    foreach (DurableOrchestrationStatus instance in result.DurableOrchestrationState)
    {
        log.LogInformation(JsonConvert.SerializeObject(instance));
    }
}

const df = require("durable-functions");

module.exports = async function(context, req) {
    const client = df.getClient(context);

    const runtimeStatus = [
        df.OrchestrationRuntimeStatus.Completed,
        df.OrchestrationRuntimeStatus.Running,
    ];
    const instances = await client.getStatusBy(
        new Date(2021, 3, 10, 10, 1, 0),
        new Date(2021, 3, 10, 10, 23, 59),
        runtimeStatus
    );
    instances.forEach((instance) => {
        context.log(JSON.stringify(instance));
    });
};

See Start instances for the function.json configuration.

import logging
from datetime import datetime
import json
import azure.functions as func
import azure.durable_functions as df
from azure.durable_functions.models.OrchestrationRuntimeStatus import OrchestrationRuntimeStatus

async def main(req: func.HttpRequest, starter: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)

    runtime_status = [OrchestrationRuntimeStatus.Completed, OrchestrationRuntimeStatus.Running]

    instances = await client.get_status_by(
        datetime(2021, 3, 10, 10, 1, 0),
        datetime(2021, 3, 10, 10, 23, 59),
        runtime_status
    )

    for instance in instances:
        logging.log(json.dumps(instance))

@FunctionName("GetRunningInstances")
public String getRunningInstances(
        @HttpTrigger(name = "req", methods = {HttpMethod.GET}) HttpRequestMessage<?> req,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext) {
    DurableTaskClient client = durableContext.getClient();
    OrchestrationStatusQuery filter = new OrchestrationStatusQuery()
        .setRuntimeStatusList(List.of(OrchestrationRuntimeStatus.PENDING, OrchestrationRuntimeStatus.RUNNING))
        .setCreatedTimeFrom(Instant.now().minus(Duration.ofDays(7)))
        .setCreatedTimeTo(Instant.now().minus(Duration.ofDays(1)));
    OrchestrationStatusQueryResult result = client.queryInstances(filter);
    return "Found " + result.getOrchestrationState().size() + " orchestrations.";
}

using Microsoft.DurableTask.Client;

// Get running or pending instances created in the last 7 days
var query = new OrchestrationQuery
{
    Statuses = new[] { OrchestrationRuntimeStatus.Running, OrchestrationRuntimeStatus.Pending },
    CreatedFrom = DateTime.UtcNow.AddDays(-7),
    CreatedTo = DateTime.UtcNow.AddDays(-1),
    PageSize = 100
};

AsyncPageable<OrchestrationMetadata> instances = client.GetAllInstancesAsync(query);

await foreach (OrchestrationMetadata instance in instances)
{
    Console.WriteLine($"{instance.InstanceId}: {instance.RuntimeStatus}");
}

from durabletask.azuremanaged.client import DurableTaskSchedulerClient
from datetime import datetime, timedelta

# Get running or pending instances created in the last 7 days
instances = client.list_orchestrations(
    created_time_from=datetime.utcnow() - timedelta(days=7),
    created_time_to=datetime.utcnow() - timedelta(days=1),
    runtime_status=['RUNNING', 'PENDING']
)

for instance in instances:
    print(f"{instance.instance_id}: {instance.runtime_status}")

import com.microsoft.durabletask.DurableTaskClient;
import com.microsoft.durabletask.OrchestrationStatusQuery;
import com.microsoft.durabletask.OrchestrationRuntimeStatus;
import java.time.Duration;
import java.time.Instant;
import java.util.List;

// Get running or pending instances created in the last 7 days
OrchestrationStatusQuery filter = new OrchestrationStatusQuery()
    .setRuntimeStatusList(List.of(OrchestrationRuntimeStatus.PENDING, OrchestrationRuntimeStatus.RUNNING))
    .setCreatedTimeFrom(Instant.now().minus(Duration.ofDays(7)))
    .setCreatedTimeTo(Instant.now().minus(Duration.ofDays(1)));

OrchestrationStatusQueryResult result = client.queryInstances(filter);

for (OrchestrationMetadata instance : result.getOrchestrationState()) {
    System.out.println(instance.getInstanceId() + ": " + instance.getRuntimeStatus());
}

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";
import { OrchestrationStatus } from "@microsoft/durabletask-js";

const client = createAzureManagedClient(connectionString);

// Get running or pending instances created in the last 7 days
const sevenDaysAgo = new Date(Date.now() - 7 * 24 * 60 * 60 * 1000);
const oneDayAgo = new Date(Date.now() - 24 * 60 * 60 * 1000);

const instances = client.getAllInstances({
    statuses: [OrchestrationStatus.RUNNING, OrchestrationStatus.PENDING],
    createdFrom: sevenDaysAgo,
    createdTo: oneDayAgo,
});

for await (const instance of instances) {
    console.log(`${instance.instanceId}: ${instance.runtimeStatus}`);
}

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

[FunctionName("TerminateInstance")]
public static Task Run(
    [DurableClient] IDurableOrchestrationClient client,
    [QueueTrigger("terminate-queue")] string instanceId)
{
    string reason = "Found a bug";
    return client.TerminateAsync(instanceId, reason);
}

const df = require("durable-functions");

module.exports = async function(context, instanceId) {
    const client = df.getClient(context);

    const reason = "Found a bug";
    return client.terminate(instanceId, reason);
};

See Start instances for the function.json configuration.

import azure.functions as func
import azure.durable_functions as df

async def main(req: func.HttpRequest, starter: str, instance_id: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)

    reason = "Found a bug"
    return await client.terminate(instance_id, reason)

param($Request, $TriggerMetadata)

# Get instance id from body
$InstanceId = $Request.Body.InstanceId
$Reason = 'Found a bug'

Stop-DurableOrchestration -InstanceId $InstanceId -Reason $Reason

@FunctionName("TerminateInstance")
public void terminateInstance(
        @HttpTrigger(name = "req", methods = {HttpMethod.POST}) HttpRequestMessage<String> req,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext) {
    String instanceID = req.getBody();
    String reason = "Found a bug";
    durableContext.getClient().terminate(instanceID, reason);
}

using Microsoft.DurableTask.Client;

string reason = "Found a bug";
await client.TerminateInstanceAsync(instanceId, reason);

from durabletask.azuremanaged.client import DurableTaskSchedulerClient

reason = "Found a bug"
client.terminate_orchestration(instance_id, reason=reason)

import com.microsoft.durabletask.DurableTaskClient;

String reason = "Found a bug";
client.terminate(instanceId, reason);

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

const reason = "Found a bug";
await client.terminateOrchestration(instanceId, reason);

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

[FunctionName("SuspendResumeInstance")]
public static async Task Run(
    [DurableClient] IDurableOrchestrationClient client,
    [QueueTrigger("suspend-resume-queue")] string instanceId)
{
    // To suspend an orchestration
    string suspendReason = "Need to pause workflow";
    await client.SuspendAsync(instanceId, suspendReason);

    // To resume an orchestration
    string resumeReason = "Continue workflow";
    await client.ResumeAsync(instanceId, resumeReason);
}

const df = require("durable-functions");

module.exports = async function(context, instanceId) {
    const client = df.getClient(context);

    // To suspend an orchestration
    const suspendReason = "Need to pause workflow";
    await client.suspend(instanceId, suspendReason);

    // To resume an orchestration
    const resumeReason = "Continue workflow";
    await client.resume(instanceId, resumeReason);
};

import azure.functions as func
import azure.durable_functions as df
from datetime import timedelta

async def main(req: func.HttpRequest, starter: str, instance_id: str):
    client = df.DurableOrchestrationClient(starter)

    # To suspend an orchestration
    suspend_reason = "Need to pause workflow"
    await client.suspend(instance_id, suspend_reason)

    # To resume an orchestration
    resume_reason = "Continue workflow"
    await client.resume(instance_id, resume_reason)

param($Request, $TriggerMetadata)

$InstanceId = $Request.Body.InstanceId

# To suspend an orchestration
$SuspendReason = 'Need to pause workflow'
Suspend-DurableOrchestration -InstanceId $InstanceId -Reason $SuspendReason

# To resume an orchestration
$ResumeReason = 'Continue workflow'
Resume-DurableOrchestration -InstanceId $InstanceId -Reason $ResumeReason

@FunctionName("SuspendResumeInstance")
public void suspendResumeInstance(
        @HttpTrigger(name = "req", methods = {HttpMethod.POST}) HttpRequestMessage<String> req,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext) {
    String instanceID = req.getBody();
    DurableTaskClient client = durableContext.getClient();  

    // To suspend an orchestration
    String suspendReason = "Need to pause workflow";
    client.suspendInstance(instanceID, suspendReason);

    // To resume an orchestration
    String resumeReason = "Continue workflow";
    client.resumeInstance(instanceID, resumeReason);
}

using Microsoft.DurableTask.Client;

// To suspend an orchestration
string suspendReason = "Need to pause workflow";
await client.SuspendInstanceAsync(instanceId, suspendReason);

// To resume an orchestration
string resumeReason = "Continue workflow";
await client.ResumeInstanceAsync(instanceId, resumeReason);

from durabletask.azuremanaged.client import DurableTaskSchedulerClient

# To suspend an orchestration
suspend_reason = "Need to pause workflow"
client.suspend_orchestration(instance_id, reason=suspend_reason)

# To resume an orchestration
resume_reason = "Continue workflow"
client.resume_orchestration(instance_id, reason=resume_reason)

import com.microsoft.durabletask.DurableTaskClient;

// To suspend an orchestration
String suspendReason = "Need to pause workflow";
client.suspendInstance(instanceId, suspendReason);

// To resume an orchestration
String resumeReason = "Continue workflow";
client.resumeInstance(instanceId, resumeReason);

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

// To suspend an orchestration
await client.suspendOrchestration(instanceId);

// To resume an orchestration
await client.resumeOrchestration(instanceId);

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

[FunctionName("RaiseEvent")]
public static Task Run(
    [DurableClient] IDurableOrchestrationClient client,
    [QueueTrigger("event-queue")] string instanceId)
{
    int[] eventData = new int[] { 1, 2, 3 };
    return client.RaiseEventAsync(instanceId, "MyEvent", eventData);
}

const df = require("durable-functions");

module.exports = async function(context, instanceId) {
    const client = df.getClient(context);

    const eventData = [ 1, 2, 3 ];
    return client.raiseEvent(instanceId, "MyEvent", eventData);
};

See Start instances for the function.json configuration.

import azure.functions as func
import azure.durable_functions as df

async def main(req: func.HttpRequest, starter: str, instance_id: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)

    event_data = [1, 2 ,3]
    return await client.raise_event(instance_id, 'MyEvent', event_data)

param($Request, $TriggerMetadata)

# Get instance id from body
$InstanceId = $Request.Body.InstanceId
$EventName = 'MyEvent'
$EventData = @(1,2,3)

Send-DurableExternalEvent -InstanceId $InstanceId -EventName $EventName -EventData $EventData

@FunctionName("RaiseEvent")
public void raiseEvent(
        @HttpTrigger(name = "req", methods = {HttpMethod.POST}) HttpRequestMessage<String> req,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext) {
    String instanceID = req.getBody();
    String eventName = "MyEvent";
    int[] eventData = { 1, 2, 3 };
    durableContext.getClient().raiseEvent(instanceID, eventName, eventData);
}

using Microsoft.DurableTask.Client;

int[] eventData = new int[] { 1, 2, 3 };
await client.RaiseEventAsync(instanceId, "MyEvent", eventData);

from durabletask.azuremanaged.client import DurableTaskSchedulerClient

event_data = [1, 2, 3]
client.raise_orchestration_event(instance_id, "MyEvent", event_data)

import com.microsoft.durabletask.DurableTaskClient;

int[] eventData = { 1, 2, 3 };
client.raiseEvent(instanceId, "MyEvent", eventData);

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

const eventData = [1, 2, 3];
await client.raiseOrchestrationEvent(instanceId, "MyEvent", eventData);

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

// Copyright (c) .NET Foundation. All rights reserved.
// Licensed under the MIT License. See LICENSE in the project root for license information.

using System;
using System.Net.Http;
using System.Threading.Tasks;
using Microsoft.Azure.WebJobs;
using Microsoft.Azure.WebJobs.Extensions.DurableTask;
using Microsoft.Azure.WebJobs.Extensions.Http;
using Microsoft.Extensions.Logging;

namespace VSSample
{
    public static class HttpSyncStart
    {
        private const string Timeout = "timeout";
        private const string RetryInterval = "retryInterval";

        [FunctionName("HttpSyncStart")]
        public static async Task<HttpResponseMessage> Run(
            [HttpTrigger(AuthorizationLevel.Function, methods: "post", Route = "orchestrators/{functionName}/wait")]
            HttpRequestMessage req,
            [DurableClient] IDurableOrchestrationClient starter,
            string functionName,
            ILogger log)
        {
            // Function input comes from the request content.
            object eventData = await req.Content.ReadAsAsync<object>();
            string instanceId = await starter.StartNewAsync(functionName, eventData);

            log.LogInformation($"Started orchestration with ID = '{instanceId}'.");

            TimeSpan timeout = GetTimeSpan(req, Timeout) ?? TimeSpan.FromSeconds(30);
            TimeSpan retryInterval = GetTimeSpan(req, RetryInterval) ?? TimeSpan.FromSeconds(1);
            
            return await starter.WaitForCompletionOrCreateCheckStatusResponseAsync(
                req,
                instanceId,
                timeout,
                retryInterval);
        }

        private static TimeSpan? GetTimeSpan(HttpRequestMessage request, string queryParameterName)
        {
            string queryParameterStringValue = request.RequestUri.ParseQueryString()[queryParameterName];
            if (string.IsNullOrEmpty(queryParameterStringValue))
            {
                return null;
            }

            return TimeSpan.FromSeconds(double.Parse(queryParameterStringValue));
        }
    }
}

const df = require("durable-functions");

const timeout = "timeout";
const retryInterval = "retryInterval";

module.exports = async function (context, req) {
    const client = df.getClient(context);
    const instanceId = await client.startNew(req.params.functionName, undefined, req.body);

    context.log(`Started orchestration with ID = '${instanceId}'.`);

    const timeoutInMilliseconds = getTimeInSeconds(req, timeout) || 30000;
    const retryIntervalInMilliseconds = getTimeInSeconds(req, retryInterval) || 1000;

    const response = client.waitForCompletionOrCreateCheckStatusResponse(
        context.bindingData.req,
        instanceId,
        timeoutInMilliseconds,
        retryIntervalInMilliseconds
    );
    return response;
};

function getTimeInSeconds(req, queryParameterName) {
    const queryValue = req.query[queryParameterName];
    return queryValue
        ? queryValue * 1000 // expected to be in seconds
        : undefined;
}

See Start instances for the function.json configuration.

import logging
import azure.functions as func
import azure.durable_functions as df

timeout = "timeout"
retry_interval = "retryInterval"

async def main(req: func.HttpRequest, starter: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)

    instance_id = await client.start_new(req.route_params['functionName'], None, req.get_body())
    logging.log(f"Started orchestration with ID = '${instance_id}'.")

    timeout_in_milliseconds = get_time_in_seconds(req, timeout)
    timeout_in_milliseconds = timeout_in_milliseconds if timeout_in_milliseconds != None else 30000
    retry_interval_in_milliseconds = get_time_in_seconds(req, retry_interval)
    retry_interval_in_milliseconds = retry_interval_in_milliseconds if retry_interval_in_milliseconds != None else 1000

    return await client.wait_for_completion_or_create_check_status_response(
        req,
        instance_id,
        timeout_in_milliseconds,
        retry_interval_in_milliseconds
    )

def get_time_in_seconds(req: func.HttpRequest, query_parameter_name: str):
    query_value = req.params.get(query_parameter_name)
    return query_value if query_value != None else 1000

Java doesn't currently have a single method for this scenario, but you can implement it with a few extra lines of code.

@FunctionName("HttpStartAndWait")
public HttpResponseMessage httpStartAndWait(
        @HttpTrigger(name = "req", route = "orchestrators/{functionName}/wait", methods = {HttpMethod.POST}) HttpRequestMessage<?> req,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext,
        @BindingName("functionName") String functionName,
        final ExecutionContext context) {

    DurableTaskClient client = durableContext.getClient();
    String instanceId = client.scheduleNewOrchestrationInstance(functionName);
    context.getLogger().info("Created new Java orchestration with instance ID = " + instanceId);
    try {
        String timeoutString = req.getQueryParameters().get("timeout");
        Integer timeoutInSeconds = Integer.parseInt(timeoutString);
        OrchestrationMetadata orchestration = client.waitForInstanceCompletion(
                instanceId,
                Duration.ofSeconds(timeoutInSeconds),
                true /* getInputsAndOutputs */);
        return req.createResponseBuilder(HttpStatus.OK)
                .body(orchestration.getSerializedOutput())
                .header("Content-Type", "application/json")
                .build();
    } catch (TimeoutException timeoutEx) {
        // timeout expired - return a 202 response
        return durableContext.createCheckStatusResponse(req, instanceId);
    }
}

using Microsoft.DurableTask.Client;

// Wait for orchestration to complete with a timeout
OrchestrationMetadata metadata = await client.WaitForInstanceCompletionAsync(
    instanceId,
    timeout: TimeSpan.FromSeconds(30),
    getInputsAndOutputs: true);

if (metadata.RuntimeStatus == OrchestrationRuntimeStatus.Completed)
{
    Console.WriteLine($"Output: {metadata.SerializedOutput}");
}

from durabletask.azuremanaged.client import DurableTaskSchedulerClient

# Wait for orchestration to complete with a timeout
state = client.wait_for_orchestration_completion(
    instance_id,
    timeout=30,
    fetch_payloads=True)

if state.runtime_status == 'COMPLETED':
    print(f"Output: {state.serialized_output}")

import com.microsoft.durabletask.DurableTaskClient;
import com.microsoft.durabletask.OrchestrationMetadata;
import java.time.Duration;
import java.util.concurrent.TimeoutException;

// Wait for orchestration to complete with a timeout
try {
    OrchestrationMetadata metadata = client.waitForInstanceCompletion(
            instanceId,
            Duration.ofSeconds(30),
            true /* getInputsAndOutputs */);
    System.out.println("Output: " + metadata.getSerializedOutput());
} catch (TimeoutException e) {
    System.out.println("Orchestration did not complete within timeout");
}

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";
import { OrchestrationStatus } from "@microsoft/durabletask-js";

const client = createAzureManagedClient(connectionString);

// Wait for orchestration to complete with a timeout
const state = await client.waitForOrchestrationCompletion(instanceId, true, 30);

if (state?.runtimeStatus === OrchestrationStatus.COMPLETED) {
    console.log(`Output: ${state.serializedOutput}`);
}

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

[FunctionName("SendInstanceInfo")]
public static void SendInstanceInfo(
    [ActivityTrigger] IDurableActivityContext ctx,
    [DurableClient] IDurableOrchestrationClient client,
    [CosmosDB(
        databaseName: "MonitorDB",
        containerName: "HttpManagementPayloads",
        Connection = "CosmosDBConnectionSetting")]out dynamic document)
{
    HttpManagementPayload payload = client.CreateHttpManagementPayload(ctx.InstanceId);

    // send the payload to Azure Cosmos DB
    document = new { Payload = payload, id = ctx.InstanceId };
}

const df = require("durable-functions");

modules.exports = async function(context, ctx) {
    const client = df.getClient(context);

    const payload = client.createHttpManagementPayload(ctx.instanceId);

    // send the payload to Azure Cosmos DB
    context.bindings.document = JSON.stringify({
        id: ctx.instanceId,
        payload,
    });
};

See Start instances for the function.json configuration.

import azure.functions as func
import azure.durable_functions as df

async def main(req: func.HttpRequest, starter: str, instance_id: str) -> func.cosmosdb.cdb.Document:
    client = df.DurableOrchestrationClient(starter)

    payload = client.create_check_status_response(req, instance_id).get_body().decode()

    return func.cosmosdb.CosmosDBConverter.encode({
        id: instance_id,
        payload: payload
    })

using namespace System.Net

param($Request, $TriggerMetadata)

$InstanceId = $Request.Body.InstanceId
$Response = New-DurableOrchestrationCheckStatusResponse -Request $Request -InstanceId $InstanceId
Push-OutputBinding -Name Response -Value $Response

[FunctionName("RewindInstance")]
public static Task Run(
    [DurableClient] IDurableOrchestrationClient client,
    [QueueTrigger("rewind-queue")] string instanceId)
{
    string reason = "Orchestrator failed and needs to be revived.";
    return client.RewindAsync(instanceId, reason);
}

const df = require("durable-functions");

module.exports = async function(context, instanceId) {
    const client = df.getClient(context);

    const reason = "Orchestrator failed and needs to be revived.";
    return client.rewind(instanceId, reason);
};

See Start instances for the function.json configuration.

using Microsoft.DurableTask.Client;

string reason = "Orchestrator failed and needs to be revived.";
await client.RewindInstanceAsync(instanceId, reason);

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

const reason = "Orchestrator failed and needs to be revived.";
await client.rewindInstance(instanceId, reason);

This sample is shown for .NET and JavaScript only.

This sample is shown for .NET and JavaScript only.

This sample is shown for .NET and JavaScript only.

[FunctionName("RestartInstance")]
public static Task Run(
    [DurableClient] IDurableOrchestrationClient client,
    [QueueTrigger("restart-queue")] string instanceId)
{
    return client.RestartAsync(instanceId, restartWithNewInstanceId: true);
}

using Microsoft.DurableTask.Client;

// Restart an orchestration with a new instance ID
string newInstanceId = await client.RestartInstanceAsync(instanceId, restartWithNewInstanceId: true);
Console.WriteLine($"Restarted as new instance: {newInstanceId}");

// Restart an orchestration keeping the same instance ID
await client.RestartInstanceAsync(instanceId, restartWithNewInstanceId: false);

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

// Restart an orchestration with a new instance ID
const newInstanceId = await client.restartOrchestration(instanceId, true);
console.log(`Restarted as new instance: ${newInstanceId}`);

// Restart an orchestration keeping the same instance ID
await client.restartOrchestration(instanceId, false);

This sample is shown for .NET and JavaScript only.

This sample is shown for .NET and JavaScript only.

This sample is shown for .NET and JavaScript only.

[FunctionName("PurgeInstanceHistory")]
public static Task Run(
    [DurableClient] IDurableOrchestrationClient client,
    [QueueTrigger("purge-queue")] string instanceId)
{
    return client.PurgeInstanceHistoryAsync(instanceId);
}

const df = require("durable-functions");

module.exports = async function(context, instanceId) {
    const client = df.getClient(context);
    return client.purgeInstanceHistory(instanceId);
};

See Start instances for the function.json configuration.

import azure.functions as func
import azure.durable_functions as df

async def main(req: func.HttpRequest, starter: str, instance_id: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)

    return await client.purge_instance_history(instance_id)

@FunctionName("PurgeInstance")
public HttpResponseMessage purgeInstance(
        @HttpTrigger(name = "req", methods = {HttpMethod.POST}, route = "purge/{instanceID}") HttpRequestMessage<?> req,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext,
        @BindingName("instanceID") String instanceID) {
    PurgeResult result = durableContext.getClient().purgeInstance(instanceID);
    if (result.getDeletedInstanceCount() == 0) {
        return req.createResponseBuilder(HttpStatus.NOT_FOUND)
                .body("No completed instance with ID '" + instanceID + "' was found!")
                .build();
    } else {
        return req.createResponseBuilder(HttpStatus.OK)
                .body("Successfully purged data for " + instanceID)
                .build();
    }
}

using Microsoft.DurableTask.Client;

// Purge a single orchestration instance
PurgeResult result = await client.PurgeInstanceAsync(instanceId);
Console.WriteLine($"Purged {result.PurgedInstanceCount} instance(s).");

from durabletask.azuremanaged.client import DurableTaskSchedulerClient

# Purge a single orchestration instance
result = client.purge_orchestration(instance_id)
print(f"Purged {result.deleted_instance_count} instance(s).")

import com.microsoft.durabletask.DurableTaskClient;
import com.microsoft.durabletask.PurgeResult;

// Purge a single orchestration instance
PurgeResult result = client.purgeInstance(instanceId);
System.out.println("Purged " + result.getDeletedInstanceCount() + " instance(s).");

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";

const client = createAzureManagedClient(connectionString);

// Purge a single orchestration instance
const result = await client.purgeOrchestration(instanceId);
console.log(`Purged ${result?.deletedInstanceCount ?? 0} instance(s).`);

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.

[FunctionName("PurgeInstanceHistory")]
public static Task Run(
    [DurableClient] IDurableOrchestrationClient client,
    [TimerTrigger("0 0 12 * * *")] TimerInfo myTimer)
{
    return client.PurgeInstanceHistoryAsync(
        DateTime.MinValue,
        DateTime.UtcNow.AddDays(-30),  
        new List<OrchestrationStatus>
        {
            OrchestrationStatus.Completed
        });
}

The purgeInstanceHistoryBy method can be used to conditionally purge instance history for multiple instances.

function.json

{
  "bindings": [
    {
      "schedule": "0 0 12 * * *",
      "name": "myTimer",
      "type": "timerTrigger",
      "direction": "in"
    },
    {
      "name": "starter",
      "type": "durableClient",
      "direction": "in"
    }
  ],
  "disabled": false
}

index.js

const df = require("durable-functions");

module.exports = async function (context, myTimer) {
    const client = df.getClient(context);
    const createdTimeFrom = new Date(0);
    const createdTimeTo = new Date().setDate(today.getDate() - 30);
    const runtimeStatuses = [ df.OrchestrationRuntimeStatus.Completed ];
    return client.purgeInstanceHistoryBy(createdTimeFrom, createdTimeTo, runtimeStatuses);
};

import azure.functions as func
import azure.durable_functions as df
from azure.durable_functions.models.DurableOrchestrationStatus import OrchestrationRuntimeStatus
from datetime import datetime, timedelta

async def main(req: func.HttpRequest, starter: str, instance_id: str) -> func.HttpResponse:
    client = df.DurableOrchestrationClient(starter)
    created_time_from = datetime.min
    created_time_to = datetime.today() + timedelta(days = -30)
    runtime_statuses = [OrchestrationRuntimeStatus.Completed]

    return await client.purge_instance_history_by(created_time_from, created_time_to, runtime_statuses)

@FunctionName("PurgeInstances")
public void purgeInstances(
        @TimerTrigger(name = "purgeTimer", schedule = "0 0 12 * * *") String timerInfo,
        @DurableClientInput(name = "durableContext") DurableClientContext durableContext,
        ExecutionContext context) throws TimeoutException {
    PurgeInstanceCriteria criteria = new PurgeInstanceCriteria()
            .setCreatedTimeFrom(Instant.now().minus(Duration.ofDays(60)))
            .setCreatedTimeTo(Instant.now().minus(Duration.ofDays(30)))
            .setRuntimeStatusList(List.of(OrchestrationRuntimeStatus.COMPLETED));
    PurgeResult result = durableContext.getClient().purgeInstances(criteria);
    context.getLogger().info(String.format("Purged %d instance(s)", result.getDeletedInstanceCount()));
}

using Microsoft.DurableTask.Client;

// Purge completed instances older than 30 days
var filter = new PurgeInstancesFilter(
    CreatedFrom: DateTime.MinValue,
    CreatedTo: DateTime.UtcNow.AddDays(-30),
    Statuses: new[] { OrchestrationRuntimeStatus.Completed });

PurgeResult result = await client.PurgeAllInstancesAsync(filter);
Console.WriteLine($"Purged {result.PurgedInstanceCount} instance(s).");

from durabletask.azuremanaged.client import DurableTaskSchedulerClient
from datetime import datetime, timedelta

# Purge completed instances older than 30 days
result = client.purge_orchestrations(
    created_time_from=datetime.min,
    created_time_to=datetime.utcnow() - timedelta(days=30),
    runtime_status=['COMPLETED']
)
print(f"Purged {result.deleted_instance_count} instance(s).")

import com.microsoft.durabletask.DurableTaskClient;
import com.microsoft.durabletask.PurgeInstanceCriteria;
import com.microsoft.durabletask.PurgeResult;
import com.microsoft.durabletask.OrchestrationRuntimeStatus;
import java.time.Duration;
import java.time.Instant;
import java.util.List;

// Purge completed instances older than 30 days
PurgeInstanceCriteria criteria = new PurgeInstanceCriteria()
    .setCreatedTimeTo(Instant.now().minus(Duration.ofDays(30)))
    .setRuntimeStatusList(List.of(OrchestrationRuntimeStatus.COMPLETED));

PurgeResult result = client.purgeInstances(criteria);
System.out.println("Purged " + result.getDeletedInstanceCount() + " instance(s).");

import { createAzureManagedClient } from "@microsoft/durabletask-js-azuremanaged";
import { OrchestrationStatus, PurgeInstanceCriteria } from "@microsoft/durabletask-js";

const client = createAzureManagedClient(connectionString);

// Purge completed instances older than 30 days
const thirtyDaysAgo = new Date(Date.now() - 30 * 24 * 60 * 60 * 1000);

const criteria = new PurgeInstanceCriteria();
criteria.setCreatedTimeTo(thirtyDaysAgo);
criteria.setRuntimeStatusList([OrchestrationStatus.COMPLETED]);

const result = await client.purgeOrchestration(criteria);
console.log(`Purged ${result?.deletedInstanceCount ?? 0} instance(s).`);

The Durable Task SDK is not available for PowerShell. Use Durable Functions instead.