Open deekerno opened 10 months ago
--additional-indexers 5
. --additional-indexers 3
. @ra0x3, I haven't seen this with the QA script. However, I haven't used it enough, and only with standard settings.
The last time I saw an issue like this was when I was doing benchmarking for metering points.
cat /opt/homebrew/var/postgresql@15/postgresql.conf | grep max_connections
max_connections = 100 # (change requires restart)
I don't remember whether this is the default or whether I edited it, so check yours.
We may be creating too many pools. If that's the case we need to make sure we clone a single handle throughout the whole indexer.
Also, we could add connection monitoring to the QA script: https://github.com/FuelLabs/fuel-indexer/issues/1433
When running our QA benchmarks, I get an error with database pool timeouts.
Command:
cargo run -p fuel-indexer-benchmarks --bin qa -- --network beta-4.fuel.network --runs 10 --blocks 25000 --num-additional-indexers 3