Sean/cloudrun public preview

Directory.Packages.props

@@ -15,16 +15,18 @@
     <PackageVersion Include="Microsoft.Extensions.Logging.Console" Version="10.0.10" />
     <PackageVersion Include="Microsoft.NET.Test.Sdk" Version="18.7.0" />
     <PackageVersion Include="Microsoft.VisualStudio.Threading.Analyzers" Version="18.7.23" />
-    <PackageVersion Include="NexusRpc" Version="0.3.0" />
+    <!-- Temporalio 1.18.0 requires NexusRpc >= 0.4.0. -->
+    <PackageVersion Include="NexusRpc" Version="0.4.0" />
     <PackageVersion Include="OpenTelemetry.Exporter.OpenTelemetryProtocol" Version="1.16.0" />
     <PackageVersion Include="StyleCop.Analyzers" Version="1.2.0-beta.435" />
     <PackageVersion Include="System.CommandLine" Version="2.0.0-beta4.22272.1" />
-    <PackageVersion Include="Temporalio" Version="1.17.0" />
-    <PackageVersion Include="Temporalio.Extensions.Aws.Lambda" Version="1.17.0" />
-    <PackageVersion Include="Temporalio.Extensions.Aws.Lambda.OpenTelemetry" Version="1.17.0" />
-    <PackageVersion Include="Temporalio.Extensions.DiagnosticSource" Version="1.17.0" />
-    <PackageVersion Include="Temporalio.Extensions.Hosting" Version="1.17.0" />
-    <PackageVersion Include="Temporalio.Extensions.OpenTelemetry" Version="1.17.0" />
+    <PackageVersion Include="Temporalio" Version="1.18.0" />
+    <PackageVersion Include="Temporalio.Extensions.Aws.Lambda" Version="1.18.0" />
+    <PackageVersion Include="Temporalio.Extensions.Aws.Lambda.OpenTelemetry" Version="1.18.0" />
+    <PackageVersion Include="Temporalio.Extensions.DiagnosticSource" Version="1.18.0" />
+    <PackageVersion Include="Temporalio.Extensions.Gcp.CloudRun.OpenTelemetry" Version="1.18.0" />
+    <PackageVersion Include="Temporalio.Extensions.Hosting" Version="1.18.0" />
+    <PackageVersion Include="Temporalio.Extensions.OpenTelemetry" Version="1.18.0" />
     <PackageVersion Include="TemporalCommunity.Aspire.Hosting" Version="0.1.0" />
     <PackageVersion Include="xunit" Version="2.9.3" />
     <PackageVersion Include="xunit.console" Version="2.9.3" />

README.md

@@ -18,6 +18,7 @@ Prerequisites:
 * [AspNet](src/AspNet) - Demonstration of a generic host worker and an ASP.NET workflow starter.
 * [Bedrock](src/Bedrock) - Orchestrate a chatbot with Amazon Bedrock.
 * [ClientMtls](src/ClientMtls) - How to use client certificate authentication, e.g. for Temporal Cloud.
+* [Gcp/CloudRun/OpenTelemetry](src/Gcp/CloudRun/OpenTelemetry) - Run a continuously-polling worker in a Google Cloud Run worker pool, exporting OpenTelemetry metrics and traces to a collector sidecar.
 * [ContextPropagation](src/ContextPropagation) - Context propagation via interceptors.
 * [CounterInterceptor](src/CounterInterceptor/) - Simple Workflow and Client Interceptors example.
 * [DependencyInjection](src/DependencyInjection) - How to inject dependencies in activities and use generic hosts for workers

TemporalioSamples.sln

@@ -7,6 +7,8 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "src", "src", "{1A647B41-53D
 EndProject
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "TemporalioSamples.ActivityWorker", "src\ActivityWorker\TemporalioSamples.ActivityWorker.csproj", "{7AECC7C6-9A21-4B8A-84D9-AFC4F5840CAF}"
 EndProject
+Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "TemporalioSamples.Gcp.CloudRun.OpenTelemetry", "src\Gcp\CloudRun\OpenTelemetry\TemporalioSamples.Gcp.CloudRun.OpenTelemetry.csproj", "{E0F934F9-10A7-41A0-A85E-EE6D0E267367}"
+EndProject
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "TemporalioSamples.Tests", "tests\TemporalioSamples.Tests.csproj", "{3FA7E5DF-03B7-4586-A980-85C155B376C5}"
 EndProject
 Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "AspNet", "AspNet", "{E431D279-E02B-4670-B934-3DB9F15D8CCC}"
@@ -163,6 +165,18 @@ Global
 		{7AECC7C6-9A21-4B8A-84D9-AFC4F5840CAF}.Release|x64.Build.0 = Release|Any CPU
 		{7AECC7C6-9A21-4B8A-84D9-AFC4F5840CAF}.Release|x86.ActiveCfg = Release|Any CPU
 		{7AECC7C6-9A21-4B8A-84D9-AFC4F5840CAF}.Release|x86.Build.0 = Release|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Debug|x64.ActiveCfg = Debug|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Debug|x64.Build.0 = Debug|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Debug|x86.ActiveCfg = Debug|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Debug|x86.Build.0 = Debug|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Release|Any CPU.Build.0 = Release|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Release|x64.ActiveCfg = Release|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Release|x64.Build.0 = Release|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Release|x86.ActiveCfg = Release|Any CPU
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367}.Release|x86.Build.0 = Release|Any CPU
 		{3FA7E5DF-03B7-4586-A980-85C155B376C5}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
 		{3FA7E5DF-03B7-4586-A980-85C155B376C5}.Debug|Any CPU.Build.0 = Debug|Any CPU
 		{3FA7E5DF-03B7-4586-A980-85C155B376C5}.Debug|x64.ActiveCfg = Debug|Any CPU
@@ -805,6 +819,7 @@ Global
 	EndGlobalSection
 	GlobalSection(NestedProjects) = preSolution
 		{7AECC7C6-9A21-4B8A-84D9-AFC4F5840CAF} = {1A647B41-53D0-4638-AE5A-6630BAAE45FC}
+		{E0F934F9-10A7-41A0-A85E-EE6D0E267367} = {1A647B41-53D0-4638-AE5A-6630BAAE45FC}
 		{E431D279-E02B-4670-B934-3DB9F15D8CCC} = {1A647B41-53D0-4638-AE5A-6630BAAE45FC}
 		{31EC2647-6A5A-42D1-B7B5-02804B340726} = {E431D279-E02B-4670-B934-3DB9F15D8CCC}
 		{AFFA4143-DC28-4FBE-A33B-D6414F541EA4} = {E431D279-E02B-4670-B934-3DB9F15D8CCC}

nuget.config

@@ -0,0 +1,22 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!--
+  TEMPORARY test scaffolding. Temporalio.Extensions.Gcp.CloudRun.OpenTelemetry is not on NuGet.org
+  yet, so this restores it (for the whole solution incl. tests) from a local folder feed. Everything
+  else restores from nuget.org. Remove this file (and src/Gcp/CloudRun/OpenTelemetry/local-packages/)
+  once the package is published.
+-->
+<configuration>
+  <packageSources>
+    <clear />
+    <add key="nuget.org" value="https://api.nuget.org/v3/index.json" />
+    <add key="local" value="./src/Gcp/CloudRun/OpenTelemetry/local-packages" />
+  </packageSources>
+  <packageSourceMapping>
+    <packageSource key="nuget.org">
+      <package pattern="*" />
+    </packageSource>
+    <packageSource key="local">
+      <package pattern="Temporalio.Extensions.Gcp.CloudRun.OpenTelemetry" />
+    </packageSource>
+  </packageSourceMapping>
+</configuration>

src/Gcp/CloudRun/OpenTelemetry/.gitignore

@@ -0,0 +1,3 @@
+# Temporary test scaffolding for the unpublished GCP Cloud Run package (remove once published).
+local-packages/
+nuget.config

src/Gcp/CloudRun/OpenTelemetry/CloudRunWorkerSample.cs

@@ -0,0 +1,20 @@
+namespace TemporalioSamples.Gcp.CloudRun.OpenTelemetry;
+
+using Temporalio.Worker;
+
+/// <summary>
+/// Shared worker configuration so both the entrypoint and the tests register the same
+/// workflow and activities.
+/// </summary>
+public static class CloudRunWorkerSample
+{
+    /// <summary>
+    /// Register the sample workflow and activities on the given worker options.
+    /// </summary>
+    /// <param name="options">Worker options to configure.</param>
+    /// <returns>The same options, for chaining.</returns>
+    public static TemporalWorkerOptions ConfigureOptions(TemporalWorkerOptions options) =>
+        options.
+            AddWorkflow<GreetingWorkflow>().
+            AddActivity(GreetingActivities.SayHello);
+}

src/Gcp/CloudRun/OpenTelemetry/Dockerfile

@@ -0,0 +1,22 @@
+# syntax=docker/dockerfile:1
+#
+# Builds the Cloud Run worker image. Temporalio (with its bundled native bridge for linux) is
+# restored from NuGet; the GCP Cloud Run OpenTelemetry package is restored from the local folder
+# feed under this sample until it is published (see nuget.config / local-packages). No SDK-from-
+# source or Rust build is needed.
+#
+# Build context is the samples-dotnet repo root (so the shared Directory.*.props / global.json are
+# available):
+#   docker build -f src/Gcp/CloudRun/OpenTelemetry/Dockerfile -t <image> .
+FROM mcr.microsoft.com/dotnet/sdk:8.0 AS build
+WORKDIR /src
+COPY global.json Directory.Build.props Directory.Packages.props .editorconfig ./
+COPY src/Gcp/CloudRun/OpenTelemetry/ ./src/Gcp/CloudRun/OpenTelemetry/
+RUN dotnet publish src/Gcp/CloudRun/OpenTelemetry/TemporalioSamples.Gcp.CloudRun.OpenTelemetry.csproj -c Release -o /app
+
+FROM mcr.microsoft.com/dotnet/runtime:8.0
+RUN useradd --create-home --uid 10001 worker
+WORKDIR /app
+COPY --from=build --chown=worker:worker /app ./
+USER 10001
+ENTRYPOINT ["dotnet", "TemporalioSamples.Gcp.CloudRun.OpenTelemetry.dll"]

src/Gcp/CloudRun/OpenTelemetry/GreetingActivities.cs

@@ -0,0 +1,14 @@
+namespace TemporalioSamples.Gcp.CloudRun.OpenTelemetry;
+
+using Microsoft.Extensions.Logging;
+using Temporalio.Activities;
+
+public static class GreetingActivities
+{
+    [Activity]
+    public static string SayHello(string name)
+    {
+        ActivityExecutionContext.Current.Logger.LogInformation("SayHello activity: {Name}", name);
+        return $"Hello, {name}!";
+    }
+}

src/Gcp/CloudRun/OpenTelemetry/GreetingWorkflow.workflow.cs

@@ -0,0 +1,19 @@
+namespace TemporalioSamples.Gcp.CloudRun.OpenTelemetry;
+
+using Microsoft.Extensions.Logging;
+using Temporalio.Workflows;
+
+[Workflow]
+public class GreetingWorkflow
+{
+    [WorkflowRun]
+    public async Task<string> RunAsync(string name)
+    {
+        Workflow.Logger.LogInformation("GreetingWorkflow started: {Name}", name);
+        var result = await Workflow.ExecuteActivityAsync(
+            () => GreetingActivities.SayHello(name),
+            new() { StartToCloseTimeout = TimeSpan.FromSeconds(10) });
+        Workflow.Logger.LogInformation("GreetingWorkflow completed: {Result}", result);
+        return result;
+    }
+}

src/Gcp/CloudRun/OpenTelemetry/Program.cs

@@ -0,0 +1,75 @@
+using System.Runtime.InteropServices;
+using Microsoft.Extensions.Logging;
+using Temporalio.Client;
+using Temporalio.Common.EnvConfig;
+using Temporalio.Extensions.Gcp.CloudRun.OpenTelemetry;
+using Temporalio.Worker;
+using TemporalioSamples.Gcp.CloudRun.OpenTelemetry;
+
+// Build client connection options from environment configuration (TEMPORAL_ADDRESS,
+// TEMPORAL_NAMESPACE, TEMPORAL_API_KEY, ...). With no API key and no TLS block this connects in
+// plaintext, which is what a local dev server (reached over an ngrok TCP tunnel) needs.
+var connectOptions = ClientEnvConfig.LoadClientConnectOptions();
+connectOptions.TargetHost ??= "localhost:7233";
+
+// Send all Temporal logs to stdout so Cloud Run captures them in Cloud Logging.
+connectOptions.LoggerFactory = LoggerFactory.Create(builder =>
+    builder.
+        AddSimpleConsole(options => options.TimestampFormat = "[HH:mm:ss] ").
+        SetMinimumLevel(LogLevel.Information));
+
+var taskQueue = Environment.GetEnvironmentVariable("TEMPORAL_TASK_QUEUE") ?? "cloud-run-worker";
+
+// The --starter mode runs a single workflow (useful for kicking off work locally against the same
+// server the deployed worker polls). Applying the defaults here too propagates a trace context into
+// the workflow so the deployed worker's spans join the same distributed trace.
+if (args.Contains("--starter"))
+{
+    using var starterTelemetry = connectOptions.ApplyGoogleCloudRunOpenTelemetryDefaults();
+    var starterClient = await TemporalClient.ConnectAsync(connectOptions);
+    var greeting = await starterClient.ExecuteWorkflowAsync(
+        (GreetingWorkflow wf) => wf.RunAsync("Temporal"),
+        new($"cloud-run-worker-{Guid.NewGuid():N}", taskQueue));
+    Console.WriteLine("Workflow result: {0}", greeting);
+    await starterTelemetry.FlushAsync(TimeSpan.FromSeconds(2));
+    return;
+}
+
+// Apply the Google Cloud Run OpenTelemetry defaults: adds the tracing interceptor and configures a
+// Temporal runtime that exports Core metrics + traces over OTLP to the local collector sidecar. The
+// returned handle owns the tracer provider and is flushed on shutdown.
+using var telemetry = connectOptions.ApplyGoogleCloudRunOpenTelemetryDefaults();
+
+var client = await TemporalClient.ConnectAsync(connectOptions);
+
+using var cts = new CancellationTokenSource();
+Console.CancelKeyPress += (_, eventArgs) =>
+{
+    eventArgs.Cancel = true;
+    cts.Cancel();
+};
+
+// Cloud Run signals shutdown with SIGTERM (about 10 seconds before SIGKILL).
+using var sigterm = PosixSignalRegistration.Create(PosixSignal.SIGTERM, _ => cts.Cancel());
+
+using var worker = new TemporalWorker(
+    client, CloudRunWorkerSample.ConfigureOptions(new(taskQueue)));
+
+Console.WriteLine(
+    "Worker running: taskQueue={0} address={1} namespace={2}",
+    taskQueue,
+    connectOptions.TargetHost,
+    connectOptions.Namespace ?? "default");
+try
+{
+    await worker.ExecuteAsync(cts.Token);
+}
+catch (OperationCanceledException)
+{
+    Console.WriteLine("Worker shutting down");
+}
+
+// Flush buffered traces within the Cloud Run shutdown grace window. Core metrics are exported
+// periodically by the runtime and have no explicit flush.
+await telemetry.FlushAsync(TimeSpan.FromSeconds(2));
+Console.WriteLine("Worker stopped");

src/Gcp/CloudRun/OpenTelemetry/TemporalioSamples.Gcp.CloudRun.OpenTelemetry.csproj

@@ -0,0 +1,11 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <PackageReference Include="Temporalio.Extensions.Gcp.CloudRun.OpenTelemetry" />
+  </ItemGroup>
+
+</Project>

src/Gcp/CloudRun/OpenTelemetry/collector-config.yaml

@@ -0,0 +1,88 @@
+# Google-Built OpenTelemetry Collector config for the Cloud Run worker-pool sidecar.
+#   metrics -> Google Managed Service for Prometheus (googlemanagedprometheus)
+#   traces  -> Cloud Trace via the Telemetry API (OTLP), authenticated with the runtime SA (ADC)
+# The worker exports OTLP/gRPC to localhost:4317; the collector detects GCP resource attributes and
+# fans out. Auth uses the worker-pool service account's Application Default Credentials via the
+# googleclientauth extension (no key files).
+receivers:
+  otlp:
+    protocols:
+      grpc:
+        endpoint: localhost:4317
+
+processors:
+  # Batch traces for throughput. Do NOT add a batch processor to the cumulative-metrics pipeline: a
+  # shutdown flush could be batched with a recent periodic export of the same series and rejected as
+  # a duplicate time series.
+  batch/traces:
+    send_batch_max_size: 200
+    send_batch_size: 200
+    timeout: 5s
+  memory_limiter:
+    check_interval: 1s
+    limit_percentage: 65
+    spike_limit_percentage: 20
+  resourcedetection:
+    detectors: [gcp]
+    timeout: 10s
+  # Rename Temporal datapoint labels that collide with the target labels Google Managed Service for
+  # Prometheus injects (e.g. Temporal emits a `namespace` label).
+  transform/collision:
+    metric_statements:
+      - context: datapoint
+        statements:
+          - set(attributes["exported_location"], attributes["location"])
+          - delete_key(attributes, "location")
+          - set(attributes["exported_cluster"], attributes["cluster"])
+          - delete_key(attributes, "cluster")
+          - set(attributes["exported_namespace"], attributes["namespace"])
+          - delete_key(attributes, "namespace")
+          - set(attributes["exported_job"], attributes["job"])
+          - delete_key(attributes, "job")
+          - set(attributes["exported_instance"], attributes["instance"])
+          - delete_key(attributes, "instance")
+          - set(attributes["exported_project_id"], attributes["project_id"])
+          - delete_key(attributes, "project_id")
+  # The Telemetry API expects the Google Cloud project in gcp.project_id.
+  transform/set_project_id:
+    error_mode: ignore
+    trace_statements:
+      - set(resource.attributes["gcp.project_id"], resource.attributes["gcp.project.id"]) where resource.attributes["gcp.project.id"] != nil
+      - set(resource.attributes["gcp.project_id"], resource.attributes["cloud.account.id"]) where resource.attributes["gcp.project_id"] == nil and resource.attributes["cloud.account.id"] != nil
+
+exporters:
+  googlemanagedprometheus:
+  otlp:
+    endpoint: telemetry.googleapis.com:443
+    compression: none
+    balancer_name: pick_first
+    auth:
+      authenticator: googleclientauth
+
+extensions:
+  health_check:
+    endpoint: 0.0.0.0:13133
+  googleclientauth:
+
+service:
+  extensions:
+    - health_check
+    - googleclientauth
+  pipelines:
+    metrics:
+      receivers: [otlp]
+      processors: [memory_limiter, resourcedetection, transform/collision]
+      exporters: [googlemanagedprometheus]
+    traces:
+      receivers: [otlp]
+      processors: [memory_limiter, resourcedetection, transform/set_project_id, batch/traces]
+      exporters: [otlp]
+  telemetry:
+    metrics:
+      readers:
+        - periodic:
+            exporter:
+              otlp:
+                protocol: grpc
+                endpoint: http://localhost:4317
+                insecure: true