wutkemtt

I did not think of the standard client libraries before as a possible solution vector. I like the idea to harden the default client implementation a littlebit. Would it be...

Our developer teams used their own reconnect implementation for the nats connection due to a lack of knowledge and not read documentation. This caused the connection issue in our cluster....

The clientID would solve the reconnection issue for us, if it is mandatory, but it is not. You can connect to a NATS server without mention a clientid. The NATS...

The MaxPubInflight issue was the root cause for the flooding of our nats streaming instance. The reconnect problem is also based on a fault client implementation. So it seems, the...

Sorry, sounded more like can't be solved in NATS server. I would like to follow the client approach on the other request as this could be a way, how to...

No problem, at the moment its okay as we prepare the next rollout phase, so we learn a lot these days. Our NATS-Streaming server runs in FT mode with the...

Yes, I meant active server, the standby took over. This failover took a few minutes because of our database, which is 10GB in size and approx. 21Mio messages. So this...

Hello, any news on this? We are discovering the same issue. Our NATS-Streaming nodes are persisting millions of messages in Postgres databases, resulting in a 32GB size for each of...

The log entry is just showing "Recovering state..." Database activity is normal within that timespan, so possibly the service is recovering the raft log. Last time this timespan was different...