We faced some issues with some RPC nodes not being able to handle requests that required a db connection to state-db, and it looks like some of the connections are not getting closed properly; we need to investigate and understand better what's happening.
We suspect this happened to the first RPCs that started and received the most requests. We have increased the number of connections to the DB, which helped avoid the problem, but we must find the source of it.
We faced some issues with some RPC nodes not being able to handle requests that required a db connection to state-db, and it looks like some of the connections are not getting closed properly; we need to investigate and understand better what's happening.
We suspect this happened to the first RPCs that started and received the most requests. We have increased the number of connections to the DB, which helped avoid the problem, but we must find the source of it.