Lease renewal in a message pump looks like a simple timer problem until you ask what happens when the runtime scheduler cannot keep its promises. Daniel Marbach's latest post digs into why that assumption matters.

The scheduler is part of your message pump

TL;DR: The async model is a contract. In .NET, async methods are expected to do little CPU work, spend most of their time awaiting input/output, and make long-running CPU work explicit. Lease renewal depends on that contract. If a handler hides blocking or CPU-heavy work inside an async method, the scheduler assumptions behind renewal can...

planetgeek.ch

Whether you want to have a CancellationToken directly in your NServiceBus Handle method, or inject dozens of different services from DI (but please don't?!) the IHandleMessages<T> interface no longer stands in your way.

Convention-based message handlers

In the previous post, we talked about how NServiceBus version 10.2 can now host multiple endpoints within the same host process. Since we’re already talking about removing constraints and limitations…

particular.net

Amazon SQS. Amazon SNS. AWS Lambda. DynamoDB. OpenTelemetry. They're powerful on their own - but the real value comes from how they work together. The AWS Loan Broker Showcase demonstrates…

Think you need messages to be delivered in order? Well, you're wrong. You just need to think about the requirements differently…and think about pizza

You don't need ordered delivery

In our family it’s a tradition that you get to decide what we’ll have for dinner when it’s your birthday. On my daughter’s last birthday, she picked pizza. I took her to the nearby pizza shop to…

particular.net

Building distributed systems is about much more than sending messages. The Loan Broker Showcase demonstrates orchestration, messaging patterns, observability, cloud services, and operational concerns in a complete reference application.

Previously you could only host a single NServiceBus endpoint in an Azure Functions app. Need another endpoint? Add another Functions App. This was…frustrating.

Most systems default to JSON without ever really asking why. Mark Seemann's latest post asks whether that default deserves a second look.

Worse is better: JSON versus XML

JSON is not a good data-interchange format.

blog.ploeh.dk

We removed the requirement to use IHandleMessages&lt;T&gt; for handler methods, but we also added Roslyn code fixes to provide scaffolding for a handler that your editor would have given you via the "Implement this interface" feature.

Convention-based message handlers

In the previous post, we talked about how NServiceBus version 10.2 can now host multiple endpoints within the same host process. Since we’re already talking about removing constraints and limitations…

t.co

Have you ever just wanted to head out on the road with your laptop and become a digital nomad? Hear how @adamralph did it, touring 7 countries in 7 months. Discover the tips and tricks he learned to make the most out of the ultimate in remote working

Code on the Road

🔗Transcription 00:10 Adam Ralph So yes, it's always great to be here. Thank you for coming to this keynote. I'm sorry I'm not Hadi, but I'll do my best to fit in. He definitely owes me a beer later…

particular.net

If a customer requests more stock than you have, is that a bug, a Result failure, or a fact worth recording as its own event? Oskar Dudycz makes the case that the answer depends on whether the business cares about that outcome afterward.

Throw, Result, or neither? - Event-Driven.io

Event-Driven by Oskar Dudycz

event-driven.io

A visibility timeout on a queue message is really a lease, not a guarantee that processing finished. When a handler runs longer than that lease, another consumer can…

What if tomorrow’s code could live in the same place as today’s code? What if it all looked the same? What if we could scale it all the same way? In this video, @adamralph will show you how to embrace the future in the code we write now

Code the future, now

🔗Transcription 00:10 Adam Ralph Hi, everyone. You hear me right? Yeah, sounds like it. So thank you very much for coming to my talk. Thank you for spending the next hour with me. My name is Adam, and…

particular.net

If a vertical slice has no service layer and no repository to mock, what exactly should a test target? Milan Jovanović makes the case for testing the whole slice as a feature instead of testing it layer by layer.

How to Test Vertical Slice Architecture

The most common question I get about vertical slice architecture isn't about structure. It's "where do my tests go?" The layered-architecture testing habits…

milanjovanovic.tech

See how to monitor your NServiceBus systems with real-time graphs for queue length, throughput, retry rate, processing time, and critical time in this video

Performance insights for NServiceBus systems

Learn about monitoring NServiceBus systems, available for ServicePulse or integrated with third-party tools like Prometheus, Azure Application Insights or New Relic.

particular.net

What happens when dozens of requests hit an expired cache key at the exact same moment? Bart Wullems walks through why the naive TryGetValue pattern lets every one of them rebuild the same expensive report in parallel.

Cache stampede: when our cache turned against us

While investigating some performance issues, we ran into an ASP.NET Core API that cached a fairly expensive aggregation query for 60 seconds...

bartwullems.blogspot.com

Few software systems operate in a vacuum and nowhere is that felt more than in the healthcare industry. Integrating with external systems is a fact of life but the usual headaches that come with it don't need to be. Make it easier with #NServiceBus

A better approach to building healthcare systems

How using NServiceBus makes it easier to build healthcare systems that are more reliable and easier to evolve over time.

particular.net

Want to understand event-driven architecture on Azure? The Azure Loan Broker Showcase brings together Azure Service Bus, Azure Functions, Azure SQL, and OpenTelemetry in a complete distributed application.

In NServiceBus 10.2, you don't have to use IHandleMessages<T>, which means you can add a CancellationToken, and inject a Logger just like with an #ASPNET controller action method. Check it out!

Convention-based message handlers

In the previous post, we talked about how NServiceBus version 10.2 can now host multiple endpoints within the same host process. Since we’re already talking about removing constraints and limitations…

particular.net

Kevin Smith compares two ways to handle unrecognized enum values from an API: falling back to an Unknown case, which is simple but loses the raw string, versus a string backed value type that preserves the original value and composes well with prefix based statuses.

Enums in API Contracts: Don't Let Your Status Values Break the World

How to consume enum values over APIs in a non-breaking way, and why the string-backed value type pattern is worth reaching for before you default to a plain C# enum.

kevsoft.net

Distributed systems aren't a collection of services - they're the interactions between them. The AWS Loan Broker Showcase demonstrates how Amazon SQS, SNS, Lambda, and DynamoDB come together in a realistic business workflow - with monitoring, tracing, and operational tooling included.

We've overhauled NServiceBus to allow you to host multiple logical endpoints in a single host process, including tools to register only those handlers and sagas that belong to each endpoint, without sacrificing the ease of use that assembly scanning gave you before.

Multiple endpoint hosting

Explore how multi-endpoint hosting in NServiceBus enables isolated endpoints inside a single .NET process while preserving clear operational boundaries.

particular.net

ASP.NET Core's IExceptionHandler interface lets you register multiple handlers, each responsible for a specific exception type, with registration order determining…

If you've ever wanted to include ServiceControl and ServicePulse in a demo or presentation, now you can! Check out our documentation on how to use the portable versions of the Particular Service Platform here:

Platform Sample installation • Platform Sample

Installation instructions for using the platform sample for demo purposes.

docs.particular.net

In 30m at #VSLive, if you're asking "should we use microservices?" then you might be asking the wrong question. Join Hazel Bohon in Room T13 to finally get clarity in Microservices for Pragmatists.

Microservices for Pragmatists -- Visual Studio Live!: Training Conferences and Events for Enterprise Microsoft .NET and Azure Developers

vslive.com

Come find us at the Particular Software booth at #VSLive to dive into messaging patterns, service-oriented architecture, and all things distributed systems. We have 3 distributed systems experts waiting to chat!

You've heard of IHandleMessages<T>, right? That interface you have to implement for every single NServiceBus message handler? It's a bit limiting, don't you think? Strict interface. Can't add a CancellationToken. Can't inject a logger.

Want to know what production messaging looks like? ✅ Multiple endpoints ✅ Distributed workflows ✅ Cloud-native architecture ✅ Observability ✅ Failure handling

Sometimes running a distributed system can feel like herding cats. At first, separating functionality into separate processes gives you flexibility you didn't have before…until suddenly you have hundreds of processes that all require orchestration.

Multiple endpoint hosting

Explore how multi-endpoint hosting in NServiceBus enables isolated endpoints inside a single .NET process while preserving clear operational boundaries.

particular.net

Event storming done. Now comes the hard part. Stream, queue, or event sourcing? July 29th, Chris Simon cuts through the hype on Kafka, RabbitMQ, Azure Service Bus, AWS SNS/SQS—with practical trade-offs you can apply immediately. Choose patterns by requirements, not trends.

It's like 10,000 streams when what you need is a queue

Choose the right messaging tech for each orange sticky note. …

particular.net

If AddSource in your OpenTelemetry setup doesn't exactly match your ActivitySource name, your traces get created, tagged, and silently dropped with no error. How many teams have chased that as a mystery bug? Bart Wullems explains in this blog post.

The role of ActivitySource in OpenTelemetry for .NET

While doing some pair programming to integrate OpenTelemetry tracing to a .NET application, we had a discussion on how to use the ActivitySo...

bartwullems.blogspot.com