We have multiple Reactive Spring Boot applications (WebFlux) running in Azure Container Apps. Since they run in Azure we can see some sporadic netty DNS resolver timeouts which we try to track down. We haven't seen them on premises. One hint is that this could be a UDP trunction problem and we should enable TCP fallback for the netty DNS resolver.
Now I would like to know two things:
I tried to find it in the Github code repos but got lost.
Sources:
Reactor Netty enables by default the TCP fallback. You can see the configuration here