You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, all of the Horizon services share the same DB connection pool. In case of increased number of HTTP requests, it's possible that ingestion service will have to wait for an idle connection. We need to create a separate connection pool (with a number of open connections equal to [horizon max] - [max needed by ingestion at any given time: 2 right now?]) for ingestion service to make sure it can always proceed.
The text was updated successfully, but these errors were encountered:
Now that we have split ingesting and front-facing nodes, this isn't necessary for our cluster. It could still potentially be helpful for single machine deployments. Let's revisit if that seems to be the case.
Currently, all of the Horizon services share the same DB connection pool. In case of increased number of HTTP requests, it's possible that ingestion service will have to wait for an idle connection. We need to create a separate connection pool (with a number of open connections equal to [horizon max] - [max needed by ingestion at any given time: 2 right now?]) for ingestion service to make sure it can always proceed.
The text was updated successfully, but these errors were encountered: