Httpz – Zero-Allocation HTTP/1.1 Parser for OxCaml
#HackerNews #Httpz #ZeroAllocation #HTTP1.1 #Parser #OxCaml #OpenSource
Httpz – Zero-Allocation HTTP/1.1 Parser for OxCaml
#HackerNews #Httpz #ZeroAllocation #HTTP1.1 #Parser #OxCaml #OpenSource
I published a new #IETF draft last week for a proposed defense against #HRS (HTTP Request Smuggling) vulnerabilities in HTTP/1.1. It's intended for use between Intermediaries (eg, CDNs and reverse proxies) and origin servers. It uses TLS Exporters (or other keys in a local context) to cryptographically protect message context (eg, the equivalent of H2 and H3 stream IDs) but in a way that attackers can't influence. It is still an early-stage -00 draft so I'm looking for general interest from potential implementers.
https://datatracker.ietf.org/doc/html/draft-nygren-httpbis-http11-request-binding
HTTP/1.1 Message Binding adds new hop-by-hop header fields that are cryptographically bound to requests and responses. The keys used are negotiated out-of-band from the HTTP datastream (such as via TLS Exporters). These header fields allow endpoints to detect and mitigate desynchronization attacks, such as HTTP Request Smuggling, that exist due to datastream handling differences.
HTTP/1.1 must die: the desync endgame
https://portswigger.net/research/http1-must-die
#HackerNews #HTTP1.1 #Must #Die #desync #endgame #HTTP #Protocol #Cybersecurity #WebDevelopment
#TIL: There are two ways to trigger use of the #HTTP3 / #QUIC #protocol in #webbrowsers:
#Chromium and #Firefox #browsers always start with #HTTP1 / #HTTP2, look for the “alt-svc” header in the response and switch to HTTP3 for subsequent requests if they find it. I knew that much.
But #Safari will instead query #DNS for the "#HTTPS" record and use that as a trigger. So it can work HTTP3-only for the cost of an additional DNS query. Unfortunately, the record type isn't widely supported yet.