I’m +1 on this , with one caveat: Did you make sure that the post_copy_size is easily made overridable ? There are cases in the code (eg CacheSM) where the overridable structs are not available.
— Leif On Tue, Dec 2, 2025 at 10:29 Brian Neradt <[email protected]> wrote: > Hi [email protected], > > In this recent PR to address issues with > proxy.config.http.request_buffer_enabled when > proxy.config.http.post_copy_size was 0: > > https://github.com/apache/trafficserver/pull/12702 > > Masakazu had the suggestion of removing request_buffer_enabled in favor of > simply using post_copy_size as the single config to control request body > buffering. I think this is a good idea. It will: > > - Be more convenient for the end user since they will need only to > configure one parameter rather than two. > - Relatedly, it will remove possible incorrect accidental > misconfigurations where the copy size is made non-zero, but request > buffering isn't enabled; or buffering is enabled and copy size is 0. > - It will simplify some of the conditional logic in the implementation. > > Pursuant to this, we will need to make post_copy_size overridable since it > is currently only a global config. > > Any concerns with deprecating proxy.config.http.request_buffer_enabled? > > Thanks, > Brian > > -- > "Come to Me, all who are weary and heavy-laden, and I will > give you rest. Take My yoke upon you and learn from Me, for > I am gentle and humble in heart, and you will find rest for > your souls. For My yoke is easy and My burden is light." > > ~ Matthew 11:28-30 >
