stream: offset in publish confirmation callback

[Дмитрий Николаевич]

Hello every body.

Suppose confirmation  delivery is on.

Is it possible to get store offset of published message for stream queue ?

  

[Arnaud Cogoluègnes]

No, it is not, the offset is not available in the confirmation.

> Suppose confirmation  delivery is on.

Note confirmation is always on with the stream protocol.

[Дмитрий Николаевич]

Are there any plans to add it ?

I know KAFKA supports it, sometimes it's useful

пятница, 12 сентября 2025 г. в 08:55:45 UTC+3, Arnaud Cogoluègnes:

[Arnaud Cogoluègnes]

No plans to add it. What is the use case?

[Дмитрий Николаевич]

I wish to store message, some dictionary, and enrich it with offset value to consume it afterwards with minimal overhead.

I know rabbitmq stream supports filter (https://www.rabbitmq.com/blog/2023/10/16/stream-filtering).

пятница, 12 сентября 2025 г. в 15:16:55 UTC+3, Arnaud Cogoluègnes:

[Karl Nilsson]

Using a stream as a kv store isn't going to be particularly efficient. scanning to find a particular offset involves an index scan which can be expensive-ish for large streams.

--
You received this message because you are subscribed to the Google Groups "rabbitmq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rabbitmq-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/rabbitmq-users/514afe62-bb00-4fc1-865c-3ee7a19f09a2n%40googlegroups.com.

--

Karl Nilsson

[Дмитрий Николаевич]

our stream is restricted by volume of 150k messages.

пятница, 12 сентября 2025 г. в 18:11:06 UTC+3, Karl Nilsson:

[Дмитрий Николаевич]

In general we need storage with high perfomance append, minimal RSS and CPU usage.

I think we can reduce max volume of stream to 50k by batching.

As I know headers with offsets and timestamps are separated from data storage.

Request rate to fetch batch is 1...10 per sec.  

пятница, 12 сентября 2025 г. в 18:22:34 UTC+3, Дмитрий Николаевич: