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
store connections, we can use the chain URL as key (fmt.Sprintf("%s:%d", chainName, *port)), gRPC can probably be configured to reconnect on failures so this store should never expire
, since we have a single connection, there's no need to initialize new clients every time
⚠️we are currently lacking a service mesh that will load balance requests among the available replicas (since we use a long lived TCP connection we remain connected to a single pod)
The text was updated successfully, but these errors were encountered:
Very similar to this api-server issue: EmerisHQ/demeris-api-server#108.
I would expect to only have one gRPC connection per chain. What we're currently doing instead is:
sdk-service/grpc_cosmos_sdk_v42.go
Line 64 in a409a12
DoD:
fmt.Sprintf("%s:%d", chainName, *port)
), gRPC can probably be configured to reconnect on failures so this store should never expiresdk-service/grpc_cosmos_sdk_v42.go
Line 73 in a409a12
The text was updated successfully, but these errors were encountered: