Cannot send messages larger than 1MB using ServiceBusClient and ServiceBusMessage with premium tier

[RafaelAZV]

• Package Name: azure-servicebus
• Package Version: 7.8.2
• Operating System: Ubuntu 22.04 LTS
• Python Version: 3.10.6

Describe the bug I want to send payloads larger than 1MB using the SB premium tier. According to the docs, by using the premium tier, the message limit size is 100MB. I already configured manually my topic to the message max size of 100MB. However, when I send a message larger than 1MB using ServiceBusMessage, I get an exception with azure.servicebus.exceptions.MessageSizeExceededError: The received message (delivery-id:1, size:1113653 bytes) exceeds the limit (1048576 bytes) currently allowed on the link.

To Reproduce Steps to reproduce the behavior:

1. Create a ServiceBusClient
2. Create a payload with more than 1MB
3. Attempt to send the message

The following code can reproduce the exception I am having:

from azure.servicebus import ServiceBusClient, ServiceBusMessage

conn_str = "<conn_str>"
client = ServiceBusClient.from_connection_string(conn_str)

payload = ["a"] * 5000000
payload = "".join(payload).encode()
message = ServiceBusMessage(body=payload)

with client:
    sender = client.get_queue_sender("<topic_name>")
    sender.send_messages(message)

Expected behavior Since the limit on my ServiceBus is 100MB, I expected the message to be send successfully

Screenshots The configuration of my topic

Additional context Additional Trace:

Traceback (most recent call last): File "/home/rafael/.local/lib/python3.10/site-packages/azure/servicebus/_base_handler.py", line 402, in _do_retryable_operation return operation(**kwargs) File "/home/rafael/.local/lib/python3.10/site-packages/azure/servicebus/_servicebus_sender.py", line 276, in _send self._handler.send_message(message.message) File "/home/rafael/.local/lib/python3.10/site-packages/uamqp/client.py", line 748, in send_message raise errors.ClientMessageError(exception, info=details) uamqp.errors.ClientMessageError: ErrorCodes.LinkMessageSizeExceeded: The received message (delivery-id:1, size:1113653 bytes) exceeds the limit (1048576 bytes) currently allowed on the link.

[kashifkhan]

Hi @RafaelAZV ,

I was not able to reproduce the error on my side. I set up a premium service bus with a queue set to accept the max size message (exactly like yours) and was able to send your payload that was provided.

I did notice you mention that your topic was set up to accept that size of message and in that case you will need servicebus_client.get_topic_sender. Just want to confirm there wasnt a mix up in my understanding.

[RafaelAZV]

Hi @kashifkhan, thank you for taking a look at my issue!

I think it is also worth to mention that my SB uses partitions (4 partitions). I made a quick experiment with the code I provided using a SB without partitions, and I was able to send the payload. Is this an expected behavior?

Could you try to reproduce using a SB with 2 or more partitions?

[kashifkhan]

Hi @RafaelAZV, I was able to repro after setting up partitions. Ive asked the question to the service bus team and will respond back as soon as I hear from them.

[RafaelAZV]

@kashifkhan ok, thank you very much!

[kashifkhan]

Hi @RafaelAZV, I got word from the service bus team that sending large messages to partitioned entities is not supported at the moment, its something they will address by the time partitioned namespaces are GA. Thanks :)

[ghost]

Hi @RafaelAZV. Thank you for opening this issue and giving us the opportunity to assist. We believe that this has been addressed. If you feel that further discussion is needed, please add a comment with the text “/unresolve” to remove the “issue-addressed” label and continue the conversation.

[ghost]

Hi @RafaelAZV, since you haven’t asked that we “/unresolve” the issue, we’ll close this out. If you believe further discussion is needed, please add a comment “/unresolve” to reopen the issue.

[yvgopal]

Partitioned namespaces feature is already in GA. But sending messages larger than 1 MB is not supported yet. It will be rolled out to production by the end of this month i.e. April 2024.