Hi, I have configured a Standard Tier Azure Service Bus and created an Azure Function App with a ServiceBusTrigger. Sessions are enabled for both the bus and trigger. My configuration are as follows: "serviceBus": { "prefetchCount": 100, "messageHandlerOptions": { "maxConcurrentSessions": 8, "autoComplete": true } } There are 3 major issues that I am facing regarding ServiceBusTriggers with Azure Functions: when a ServiceBusTrigger is configured with sessions enabled and prefetch count more than 1, messages for a particular session are processed out of order after some messages fail to process initially, some of the logs for retries are not shown in the log stream, when processing for multiple sessions, messages for some sessions are received very late while others are received immediately. If for example, there are 8 sessionIDs and 8 max concurrent sessions, messages for 2 or 3 sessions are being received after a significant amount of time. This behavior is not occurring all the time. Does anyone have an idea of why this would be happening? Thankyou in advance for the help!

Hello @Ume Hani The issues you are facing could be related to how the ServiceBusTrigger is handling sessions and concurrent processing. Here are some possible reasons for the problems you described: Out of Order Processing : When using sessions with a prefetch count greater than 1, messages for a particular session may be processed out of order if some messages fail to process initially. This could be due to the way messages are being retrieved and processed concurrently. Missing Retry Logs : The missing logs for retries could be related to how the retry logic is implemented in your function code. It's important to ensure that the retry mechanism is correctly handling failed messages and logging the retry attempts. Delayed Message Reception : The delayed reception of messages for some sessions could be caused by the way concurrent sessions are being processed. If messages for some sessions are taking longer to process, it could be impacting the overall message processing time. To address these issues, you may need to review and optimize your Azure Function code that handles the ServiceBusTrigger. Make sure that the message processing logic is robust, handles retries effectively, and manages concurrent sessions efficiently. Additionally, you may consider adjusting the prefetch count, maxConcurrentSessions, and other configuration settings to better align with your processing requirements and message handling logic. If the issues persist, you may want to consider reaching out to Azure support for further assistance and troubleshooting. I hope this information helps you identify and resolve the issues you are facing with the ServiceBusTrigger configuration in Azure Functions.

Issues with ServiceBusTrigger configuration with Azure Functions

Ume Hani 0

Hi,

I have configured a Standard Tier Azure Service Bus and created an Azure Function App with a ServiceBusTrigger. Sessions are enabled for both the bus and trigger. My configuration are as follows:

"serviceBus": {
 	"prefetchCount": 100,
	"messageHandlerOptions": {
		"maxConcurrentSessions": 8,
		"autoComplete": true
	 }
 }

There are 3 major issues that I am facing regarding ServiceBusTriggers with Azure Functions:

when a ServiceBusTrigger is configured with sessions enabled and prefetch count more than 1, messages for a particular session are processed out of order after some messages fail to process initially,
some of the logs for retries are not shown in the log stream,
when processing for multiple sessions, messages for some sessions are received very late while others are received immediately. If for example, there are 8 sessionIDs and 8 max concurrent sessions, messages for 2 or 3 sessions are being received after a significant amount of time. This behavior is not occurring all the time.

Does anyone have an idea of why this would be happening? Thankyou in advance for the help!

Loknathsatyasaivarma Mahali 430 Reputation points Microsoft Vendor

2025-02-12T09:25:38.38+00:00

Hello @Ume Hani,
Thanks for posting your question on Microsoft Q&A Platform!

When a ServiceBusTrigger is configured with sessions enabled and prefetch count more than 1, messages for a particular session are processed out of order after some messages fail to process initially --> how you have confirm that messages were out of order ? if messages are sending with session id all the messages will be send in order if any of the message is not consumed by your application that will be send to dead letter queue .

And some of the logs for retries are not shown in the log stream ---> Are you logging messageid and retry count in case if you are logging these properties then you can review the application insights logs to look into particular message id to check whether the message is retried or not.

When processing for multiple sessions, messages for some sessions are received very late while others are received immediately. If for example, there are 8 sessionIDs and 8 max concurrent sessions, messages for 2 or 3 sessions are being received after a significant amount of time. This behavior is not occurring all the time. -->_Message can be consumed by any of the session id once the all the message is consumed from a particular session Id then your client application tries to consume message from other session In case of single instance of your application) in any if you have multiple instance of your application is running then the application can consume message from any of the partitions for more details how sessions works.

Also, for your reference please refer the below document for better understanding:_ Message sessions
Ume Hani 0 Reputation points

2025-02-12T10:18:52.8833333+00:00
@Loknathsatyasaivarma Mahali Let me give you a breakdown of the functionality I am trying to acheive so that you can better understand these issues:

I have created a service bus sender application through which i am sending messages with 8 different sessionIDs ( Tenant-0, Tenant-1 ... ) concurrently with 8 different sender threads, each new message has an incremental sequence number that can be used to validate the order of messages,

messages, when received, will be logged as follows:

Event: {"tenantID":"Tenant-4","producerID":4,"correlationID":"Corr-0e4f6bac-d114-4f87-8413-94a2b6dd81da","eventType":"UPDATE","tenantSequenceNumber":35}

I have a function app which has a ServiceBusTrigger that receives these messages and sends them for processing to a separate server,

the server validates whether the messages for a specific session are being received in order or not (via sequence numbers). It throws exceptions if the messages are received out of order. In the logs we are seeing these exceptions.

If the server is down, it throws an Exception in the function app and the message delivery is retried,

For validating sequences after retry, I do the following test:

run the function app,

start the sender without starting the server,

start the server after some delay so that some messages are retried,

monitor sequence mismatch errors for these messages that were retried,

The expectation is that even when a message is retried, the messages for that session should be executed in order, preserving sequence number. However, after a message is retried, the sequence of execution does not follow sequence numbers i.e. the sequence in which the messages were sent.

If we set prefetch count to 0, then the messages are received in sequence even with retries.

The messages are received in order if there are no retry attempts. However, if a message fails to deliver ( e.g when our server is down ), it gets retried and then it is received out of order.

Also, most logs for retried messages are correctly shown in application insights. However, some logs do not show up. E.g. the log for retry 1,2,3 .., 5,6 will be present, but 4 will be missing and so on.
Loknathsatyasaivarma Mahali 430 Reputation points Microsoft Vendor

2025-02-12T17:42:30.8433333+00:00
Hello @Ume Hani,

Thanks for sharing the detailed breakdown of the functionality you are trying to achieve with your service bus sender application, function app, and server setup. It seems like you are facing issues with message sequencing and retries.

To provide you with the appropriate solution could you please confirm the below for our better investigation:

Can you please tell us how the client sends the message to downstream app (is it a separate app domain because you mentioned "different server")

Also, how often do the missing logs happen? is it every "session"? every other "session"?

1 answer

Pinaki Ghatak 5,575 Reputation points Microsoft Employee

2025-02-13T17:14:53.21+00:00
Hello @Ume Hani

The issues you are facing could be related to how the ServiceBusTrigger is handling sessions and concurrent processing.

Here are some possible reasons for the problems you described:

Out of Order Processing: When using sessions with a prefetch count greater than 1, messages for a particular session may be processed out of order if some messages fail to process initially. This could be due to the way messages are being retrieved and processed concurrently.

Missing Retry Logs: The missing logs for retries could be related to how the retry logic is implemented in your function code. It's important to ensure that the retry mechanism is correctly handling failed messages and logging the retry attempts.

Delayed Message Reception: The delayed reception of messages for some sessions could be caused by the way concurrent sessions are being processed. If messages for some sessions are taking longer to process, it could be impacting the overall message processing time.

To address these issues, you may need to review and optimize your Azure Function code that handles the ServiceBusTrigger. Make sure that the message processing logic is robust, handles retries effectively, and manages concurrent sessions efficiently. Additionally, you may consider adjusting the prefetch count, maxConcurrentSessions, and other configuration settings to better align with your processing requirements and message handling logic. If the issues persist, you may want to consider reaching out to Azure support for further assistance and troubleshooting.

I hope this information helps you identify and resolve the issues you are facing with the ServiceBusTrigger configuration in Azure Functions.
Please sign in to rate this answer.

0 comments No comments
Sign in to comment

Use comments to ask for clarification, additional information, or improvements to the question.

Share via

Issues with ServiceBusTrigger configuration with Azure Functions

1 answer

Your answer