peers
ERROR – "PeerJS: " – "Error:" – "Lost connection to server."
While running my code i'm getting this error referring to peerjs in dev tools. I'm using peerjs version 1.3.1 (https://unpkg.com/[email protected]/dist/peerjs.min.js). Now I noticed that a new version (1.3.2)has just been released. I tried to use it but i get the same error. I tried to use an older version but i get the same error. Does anyone have the same problem ? Is there anybody who can help me ?
Example code : https://morgan3d.github.io/misc/jschat/
This problem is probably caused by peer-server being down. Here is the issue: #805
You can follow the thread of the issue I posted and see if any of the suggested solutions could work for you.
what do you suggest I do to solve the problem ?
how did you resolved this problem Lost connection to server
The server is up with a pretty decent track record. See our status page. If it's marked as online there and not reachable for you, most likely it's a network or DNS error on your side.
@jonasgloning
It's happening to me too. peerjs library can't reach wss://0.peerjs.com
Ping works it's not DNS problem
Server reponsd
I don't know why it happens few hours ago it worked noramlly
Looks like it's a known problem from the past I encourage you to mention in the README that the default servers are not stable and goes down from time to time. https://news.ycombinator.com/item?id=25659044
Yes, currently there's a DDoS from North America. What was a steady stream of a few hundred thousand to million connections grew to half a billion requests in the last days with weird traffic patterns. More compute is on the way, and Cloudflare is doing it's part for now.
There is a status page, also linked in the readme, with a transparent uptime over the last months. I don't think "not stable" characterises the >99,9% uptime over the last 6 months well, although I understand the frustration. The situation is also very different than 2021, with a new maintainer, new infrastructure and a different server implementation.
I'm going to lock this issue; it was opened years ago about something different. Please feel free to open a new one to discuss there further, if needed.