RFR: 8371903: HttpClient: improve handling of HTTP/2 GOAWAY frames with error code [v7]

[Daniel Fuchs]

On Thu, 11 Dec 2025 12:24:29 GMT, Jaikiran Pai <jpai at openjdk.org> wrote:

>> @jaikiran won't the code at line 1483 below (`close(cause)`) ensure that the connection is actually closed?
>
> Hello Daniel, line 1483, does guarantee that the connection will actually get closed (if it isn't already).
> In the current proposed form in this PR, the potential issue is that the processed streams are being passed a termination cause (the one from goaway frame), however the connection itself might get closed by a different cause, by the time the close on line 1483 is run. So we can end up in a situation where the connection closure cause is something else and the stream termination cause is something else.

Oh - I see - good catch then!

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/28632#discussion_r2610404534