> Am 13.05.2024 um 13:56 schrieb Eric Covener <cove...@gmail.com>:
>
> On Mon, May 13, 2024 at 7:32 AM Stefan Eissing via dev
> <dev@httpd.apache.org> wrote:
>>
>> I have merged https://github.com/icing/mod_h2/pull/280 to the mod-h2 on
>> github. With mpm_event, this will return HTTP/2 connections more often to
>> the mpm, thus freeing a worker.
>>
>> While this sounds good, I am not sure this is beneficial for a server under
>> load. The current connection state handling is designed for HTTP/1.x where a
>> connection is "given back" to the MPM at the end of a request.
>>
>> AFAICT, the MPM assumes that, once any pending output is written, it may
>> close the connection under load. Because in HTTP/1.x it means the connection
>> has served the last response completely. The client should be grateful and
>> cope well with the connection being closed subsequently due to other clients
>> demands.
>>
>> In HTTP/2 so far, we did return to the MPM only when all requests had been
>> served. The connection is therefore really in a similar state to HTTP/1.x.
>> The client has got its responses, we are free to close.
>>
>> With the change in PR 280, we return on being flow blocked. The response(s)
>> are *not* finished. If MPM now closes such a connection under load, the
>> client will most likely try again. This seems counter productive.
>>
>> Therefore I am hesitant if this change is beneficial for us. It frees up a
>> worker thread - that is good - but. Do we need a new "connection state" here?
>
> We have CONN_STATE_SUSPENDED too. These are all but forgotten by the
> MPM until something changes their state back.
> You may see some references to PT_USER that got backported in
> comments, but that is trunk-only.
> PT_USER is about how something gets help un-suspending. In the case of
> H2, I guess the non-worker thread knows/should know when it could
> resume a request already?
In this case, HTTP/2 is only waiting for a POLLIN on the connection's socket.
No progress can be made until that happens or the connections Timeout occurs.
>
> ./modules/test/mod_dialup.c is a sample-ish program that uses
> SUSPENDED to not tie up a thread while it sleeps
> In trunk, modules/proxy/mod_proxy_wstunnel.c and mod_proxy_http use it
> to avoid an idle tunnel tieing up a thread.
