Tristan Hyams

@LzyRapx A popular strategy is to pull out the message from a Channel with Confirm enabled, so you can perform Acks/Nacks like you have sample code for. Then you try...

I can submit a v2 migration to baseline if interested @streadway

You can close this ticket out, while it is nice to keep these things up to date, the version of Golang that had the issue was in the 11.x branch...

You do want to avoid using channels in parallel. I.e. many goroutines, many publishes with the same channel. In this library, I believe its safe, but its inefficient/blocking with all...

After playing around with it some more... The length prior to the select is 0 and it is in fact closed (1). So it never needs to be `nil` and...

I created channel/connection pools and have been in production. https://github.com/houseofcat/turbocookedrabbit Always room for improvement (submit PRs!) and it's based off of Streadway/amqp.

ConsumerPipelineMicroservice (R9 5950X) Global Settings With Simulated IO delay per processing each message. ```csharp public static Stopwatch Stopwatch; public static LogLevel LogLevel = LogLevel.Error; public static long GlobalCount = 200_000;...

Here is the DataProducer Single Publisher rate.  As expected - per RabbitMQ - as the queue fills up, performance decreases. This is where multiple queues (of the same data...

Consumer Stats on ConsumerPipelineMicroservice (1 Consumer Count)  Cpu On ConsumerPipelineMicroservice  AppTestSettings ```js { "HouseofCat": { "ConsumerDataflowOptions": { "DataflowName": "ConsumerDataflow", "ConsumerName": "ConsumerFromConfig", "ConsumerCount": 1, "MaxDoP": 64, "EnsureOrdered": false, "Capacity":...

Now, the above test is about allocating a ton of CPU to the Consumer. The following test demonstrates scaling out internally with more consumers with less CPU resources to achieve...