Hi. I need help. I’m setting up DominoIQ for remote use RAG (I use ollama for testing), but when i initializing the vector (updall -w), I get the error: Error: Vector host not specified.
Excuse me, what maintenance are you talking about if vectorization doesn’t complete based on the instructions?
Maintenance should be performed once vectorization is already in place, but the problem is that with the Remote RAG, vectorization DOESN’T OCCUR for some reason (
The DEBUG_DOMIQ=1 parameter is already set. And the errors reported are only displayed because of it.
There are no LLM logs in the IBM Technical Support folder. Try setting up a remote RAG yourself. What behavior will you get?
Unfortunately, the documentation doesn’t describe how to set up a remote embedded model. Maybe it still doesn’t work and I’m wasting my time researching?
I’m realy sure, RAG is currently only possible with an LLM running on the DominoIQ server, not with a remote LLM.
In Domino 14.5.1, RAG support is available only when the Domino IQ server is configured in Local mode, so the LLM handling RAG source content is running on the Domino IQ servers configured for RAG.