mathstodon.xyz is one of the many independent Mastodon servers you can use to participate in the fediverse.
A Mastodon instance for maths people. We have LaTeX rendering in the web interface!

Server stats:

2.8K
active users

#http2

0 posts0 participants0 posts today

How browsers REALLY load Web pages

When browsers load a Web page and its subresources, A LOT happens under the hood. They need to take into account render/parsing blocking resources, use a preload scanner, listen to resource hints (like preload/preconnect), loading modifiers (async/defer/module), fetchpriority, responsive images, and much more. […]

📺 fosdem.org/2025/schedule/event

fosdem.orgFOSDEM 2025 - How browsers REALLY load Web pages
#video#web#browser
Continued thread

Scheduling HTTP Streams
by Alexander Krizhanovsky.

Understand how a web server manages millions of concurrent web requests.
* Differences in treatment for progressive JPEG vs a PNG.
* Resilient against DDoS attacks and other vulnerabilities.
* Compare Nginx, Envoy, Apache.

fosdem.org/2025/schedule/event

Live stream:
live.fosdem.org/watch/ua2220

Continued thread

How browsers REALLY load Web pages by Robert Marx @programmingart

Browsers go through great lengths to workaround the fact that most web servers and major CDN provides do not honour the HTTP fetch priority correctly.

The same page has a very different waterfall in Firefox, Chrome, and Safari.

fosdem.org/2025/schedule/event

Live stream:
live.fosdem.org/watch/ua2220

I'm trying to track down a weird issue with using #streams as request body for #fetchapi requests in the #browser. After the first chunk is read, the connection stalls and the server never receives any data. It seems to be related to #http3 / #quic. When I downgrade the http reverse proxy to #http2, it works fine.

Has anyone experienced this too? Are streaming request bodies not compatible with http3? (as currently implemented in #chromium)

Replied to dusoft

@dusoft@fosstodon.org Wow, you'd think that #Chrome, which is developed by #Google which developed #SPDY which became the basis for #HTTP2, would be strict in how they process HTTP/2 response headers... 🤦‍♂️

Mozilla should just WONTFIX this. The spec clearly says they should be lowercase, Isn't one of the points of HTTP/2 is that we do away with this practice of trying to accept lazy malformed responses that became endemic in the HTTP/1 era?
​:seija_coffee:​