I've been dealing with memory issues and OOM for the LAPI pod. Fixed with chunked_decisions_stream=true. I don't see why this option should not be enabled considering the size of the intial decision fetch by bouncers.
I've been dealing with memory issues and OOM for the LAPI pod. Fixed with chunked_decisions_stream=true.
I don't see why this option should not be enabled considering the size of the intial decision fetch by bouncers.