Revisor

Results 16 comments of Revisor

Just to recap, the first connection right now occurs as follows: 1. `Disque\Connection\Manager::connect()` calls 2. `Manager::findAvailableConnection()`, which shuffles all known nodes randomly and tries to connect to the first working...

I think it is not possible to connect to a leaving node. If the connection, authentication, or HELLO return an error (and thus throw an exception), the node is skipped...

Out of curiosity, why would multiple consumers be waiting for one job to be processed?

Try changing the model from `gpt-4` to `gpt-3.5-turbo`. It is possible your account doesn't have API access to GPT4 (yet).

-1 from me, create a PR for the Bitrix adapter. It's just one file.