When handling HTTP requests, the HttpClient implementation in the JDK either creates a new HttpConnection or reuses a pooled instance for the target origin server to which this request is directed. The origin server is deduced from the request's URI. Right now the HttpConnection doesn't hold on to that origin server information.

During the review of upcoming HTTP3 support in HttpClient, it has been suggested that we should keep track of the origin server in the HttpConnection so that this information is available to the rest of the code, for example when dealing with alternate services https://github.com/openjdk/jdk/pull/24751#discussion_r2172224088.