On Fri, 2024-11-01 at 11:42 -0400, Brent Putman wrote: > Hi, > > We have a question about the exact behavior of the new > responseTimeout > setting introduced in 5.0. Its Javadocs say: "Determines the timeout > until arrival of a response from the opposite endpoint." > > To us it isn't clear what "arrival of a response" actually means > precisely. Is it when the initial bytes of the response header start > showing up? Or the complete receipt of all bytes of header + body? > Or > something else? >
Hi Brent Presently a response timeout is the maximum of period inactivity until receipt of the response message _head_. It is basically nothing more than a socket timeout. In the future versions of HttpClient semantics of this timeout may change. > Mainly we're trying to understand if this new setting implements the > same absolute limit on the total allowed request/response time as one > gets with the various approaches that have been documented in the > past > involving an async thread (via for example TimerTask or > ScheduledExecutorService) that fires the request's abort()/cancel() > when the specified time delay is reached. > Presently it does not implement an absolute limit or a deadline. This may change in the future. Oleg --------------------------------------------------------------------- To unsubscribe, e-mail: httpclient-users-unsubscr...@hc.apache.org For additional commands, e-mail: httpclient-users-h...@hc.apache.org
