Closed
Description
Description
Currently we can only set one failed queue that is available for every transport.
What happens if we want different behaviours?
Like, one queue that we just want to discard the failed queue and others where we care and want to use the failed queue (a specific one or the same global one)?
Example
# config/packages/messenger.yaml
framework:
# no need to set failed transport globally if I want a specific behaviour per transport.
messenger:
transports:
failed: 'doctrine://default?queue_name=failed'
failed_important: 'doctrine://default?queue_name=failed_important'
async:
dsn: '%env(MESSENGER_TRANSPORT_DSN)%'
failure_transport: failed
retry_strategy:
max_retries: 3
delay: 1000
multiplier: 2
async_no_failure_transport:
dsn: '%env(MESSENGER_TRANSPORT_DSN)%'
retry_strategy:
max_retries: 3
delay: 1000
multiplier: 2
async_send_specific_failure_queue:
dsn: '%env(MESSENGER_TRANSPORT_DSN)%'
failure_transport: failed_important
retry_strategy:
max_retries: 3
delay: 1000
multiplier: 2This feature covers for example cases where you need to make distinguish between:
- Events that are constantly being generated and you don't care if they fail after the number of retries
- Events that you need to be absolutely sure that they need to be executed, and moved to a different failed queue (to be debugged by a specific team related with a specific queue. It could even be a very important queue, and you need to prioritise the debug of those events)
I don't mind to try to implement this behaviour.
Just wanted to know your feedback.