How do I know if I am sending successfully after using BasicPublishAsync

[a645010545]

Is your feature request related to a problem? Please describe.

BasicPublishAsync

Describe the solution you'd like

var properties = new BasicProperties();
properties.DeliveryMode = DeliveryModes.Persistent;
await channel.BasicPublishAsync(exchange, string.Empty, true, properties, bytes);

How do I know if I am sending successfully after using BasicPublishAsync

Describe alternatives you've considered

No response

Additional context

No response

[michaelklishin]

@a645010545 this should have been a discussion.

See this example app that demonstrates how publisher confirms can be used.

[a645010545]

System.Threading. SemaphoreFullException: Adding the specified count to the semaphore would cause it to exceed its maximum count.
Why does BasicPublishAsync report this error

[michaelklishin]

Please share your code and a complete stack trace. We do not guess in this community.

[a645010545]

public static IConnection Connection { get; set; }
private static ConnectionFactory _factory;
private static IChannel _channel;

public static async Task Init()
{
    if (Connection == null)
    {
        var option = Service.Current.Reslove<IOptions<MessageMiddlewareOptions>>();
        _factory = new ConnectionFactory
        {
            HostName = option.Value.Host,
            UserName = option.Value.Username,
            Password = option.Value.Password,
            AutomaticRecoveryEnabled = true
        };
        Connection = await _factory.CreateConnectionAsync();
    }
    if (_channel == null)
    {
        var channelOpts = new CreateChannelOptions(
            publisherConfirmationsEnabled: true,
            publisherConfirmationTrackingEnabled: true,
            outstandingPublisherConfirmationsRateLimiter: new ThrottlingRateLimiter(256)
        );
        _channel = await Connection.CreateChannelAsync(channelOpts);
    }
}

public static async Task Push(string exchange, object data)
{
    int retryCount = 0;
    const int maxRetries = 3;
    while (retryCount <= maxRetries)
    {
        try
        {
            var jsonStr = JsonConvert.SerializeObject(data);
            var bytes = Encoding.UTF8.GetBytes(jsonStr);
            var properties = new BasicProperties();
            properties.DeliveryMode = DeliveryModes.Persistent;
            await _channel.BasicPublishAsync(exchange, string.Empty, true, properties, bytes);
            break;
        }
        catch (Exception ex)
        {
            retryCount++;
            Console.WriteLine(string.Format("RabbitMQ Push ERROR:{0}", ex.ToString()));
        }
    }
}

RabbitMQ Push ERROR:System.Threading. SemaphoreFullException: Adding the specified count to the semaphore would cause it to exceed its maximum count.
at System.Threading.SemaphoreSlim. Release(Int32 releaseCount)
at RabbitMQ.Client.Impl.Channel. MaybeEndPublisherConfirmationTracking(PublisherConfirmationInfo publisherConfirmationInfo, CancellationToken cancellationToken)
at RabbitMQ.Client.Impl.Channel. BasicPublishAsync[TProperties](String exchange, String routingKey, Boolean mandatory, TProperties basicProperties, ReadOnlyMemory`1 body, CancellationToken cancellationToken)
at Lixise.HealthCloud.Infrastructure.Amqp.MessageMiddleware. Push(String exchange, Object data) in D:\云平台项目\Lixise.Yun\Tcp\src\Lixise.HealthCloud.Infrastructure\Amqp\MessageMiddleware.cs:line 64

[a645010545]

Is it because of this MAX_OUTSTANDING_CONFIRMS

[lukebakken]

No, it's not. You will have to provide more information:

• What version of this library are you using?
• What does RabbitMQ log when you get SemaphoreFullException?
• Does this exception happen every single time?
• Can you log first chance exceptions to see if there is something else going on? https://learn.microsoft.com/en-us/dotnet/framework/app-domains/how-to-receive-first-chance-exception-notifications

I have never seen this scenario. We have extensive tests for publisher confirmations so this must be due to your code somehow.

You can see test application code here: https://github.com/rabbitmq/rabbitmq-dotnet-client/blob/main/projects/Applications/PublisherConfirms/PublisherConfirms.cs

If you can provide a complete code sample that I can clone, compile, and run, to observe this issue, I can quickly diagnose it. Thanks.

[lukebakken]

My guess is that SemaphoreFullException may be masking another exception.

[lukebakken]

@a645010545 I wonder if your retry logic causes this exception to be thrown - https://github.com/rabbitmq/rabbitmq-dotnet-client/blob/main/projects/RabbitMQ.Client/Impl/Channel.PublisherConfirms.cs#L289

Can you check?

[a645010545]

The error won't occur if you delete this parameter. outstandingPublisherConfirmationsRateLimiter: new ThrottlingRateLimiter(256)

[lukebakken]

Please, take the time to read my comments and the questions I asked, and answer them.

I have added your code to a test in this PR, and I do not see the same issue as you.

So, there must be something different about what you are doing that causes the issue you report. I don't have time to guess, so please provide more information.

[lukebakken]

@a645010545 - I don't know why you are running into a case where your ThrottlingRateLimiter(256) instance can't acquire a lease, but that is most likely the cause of the SemaphoreFullException.

See this PR, which reproduces (artificially) this issue, and fixes it: #1794

[a645010545]

Try it with this demo?MaybeSomethingUpWithRateLimiter_GH1793

[a645010545]

Adding the specified count to the semaphore would cause it to exceed its maximum count.?