Lack of Websocket Support in Xray Core for Iranian Users

[masood5400]

Hello,

I hope this message finds you well. I would like to address an issue regarding the Xray Core repository. Firstly, I want to express my appreciation for the fantastic new Reality feature that works seamlessly on TCP and gRPC. It has greatly improved my experience.

However, I would like to bring to your attention the fact that Xray Core currently lacks support for Websocket connections. This limitation poses a challenge for users in Iran who rely on Websocket for certain applications and integrations. I understand that implementing this feature may require significant development effort, but I believe it would greatly enhance the usability of Xray Core for a broader audience.

Could you please provide an estimated timeline for when Websocket support might be available in Xray Core? This information would be extremely helpful for planning purposes and managing user expectations.

Thank you for your attention to this matter. I look forward to your response and the future improvements to Xray Core.

Best regards

[FranzKafkaYu]

Could you please provide an estimated timeline for when Websocket support might be available in Xray Core?

websocket is available in xray core long time ago,plz follow this link and check more details.

[lostsoul6]

Could you please provide an estimated timeline for when Websocket support might be available in Xray Core?

websocket is available in xray core long time ago,plz follow this link and check more details.

I think he meant websocket with REALITY .

[masood5400]

I mean websocket with REALITY .

[us254]

https://github.com/XTLS/Xray-core/issues/1750#issuecomment-1459979602

WebSocket is a protocol that has been designed to be extremely failed, and it has a complex and redundant set of protocol structures and state control of its own, but does not come with multiplexing, but instead has to shake hands one more time each time.

[us254]

https://github.com/XTLS/Xray-core/issues/1750#issuecomment-1459469821

By the way, let me explain why WSS Proxy is accurately identified:

Fingerprint: Even if the camouflage is opened, the ALPN it issues is always , and at a glance WSS, it actually cannot do what we want to "hide wood in the forest", and will only give away people's heads naked. Handshake: WS in the WSS layer has to shake hands more than once, and the timing characteristics are very unique. In fact, opening early data can be alleviated, if you have to use WSS, it is recommended ?ed=2048 TLS in TLS: This is a common feature of TLS proxies and requires targeted processing. Multiplexing can alleviate the inner TLS handshake feature, but it aggravates the characteristics of the "cryptographic matryoshka doll", reference XTLS Vision, fixes TLS in TLS, to the star and beyond #1295 The second paragraph, so currently XTLS Vision is the better solution.

So my advice now is: don't use WSS, and it should be listed as deprecated. The CDN has gRPC, and the direct connection has N postures, and there is no need to use WSS.

[us254]

https://github.com/XTLS/Xray-core/discussions/1921#discussioncomment-5562063

Reality, working principle is different. I think it can never use cdn. unless it discards its original function and redesigns it. If its functions and disadvantages become the same as ws+cdn, then it makes no sense.

[RPRX]

REALITY WebSocket 这个组合没有任何必要，毕竟你不能用它过 CDN，且直连更不必用它。