Rémy, On 9/28/15 3:33 AM, Rémy Maucherat wrote: > 2015-09-27 14:32 GMT+02:00 Christopher Schultz <ch...@christopherschultz.net >> : > >> >> In general, I think that the proposal to abandon AJP and use HTTP/2 >> instead makes a lot of sense. However, AJP still does offer an advantage >> over HTTP/1 or HTTP/2: it allows information to be passed out-of-band >> with respect to the message itself that can be considered trusted. >> >> Obviously HTTP/1.1 is not extensible, and manipulating headers is a mess > there. That's not the case with HTTP/2, for example it is possible to add > an extra custom frame type to transfer all our "proxy stuff". Obviously it > needs to be thought about a bit more than that ...
I didn't realize that HTTP/2 allowed for custom frame types like that. It seems a natural candidate for that "proxy stuff" :) > Of course, if that AJP update is trivial, then why not do it as well, but > since many people are using HTTP for proxying, combined with the length of > time it will take for an actual "AJP+" adoption, I'm not convinced it will > be used a lot. Agreed. The advantages of AJP are dwindling: it's better to focus on a protocol that is more flexible and covers more use-cases. -chris
signature.asc
Description: OpenPGP digital signature
